CAUTION!
The material in this book is for educational
and informational purposes only. The projects
contained in this book are not suitable for children
without close supervision by a knowledgeable adult.
CAUTION!
Fuel cell systems involve hydrogen and air, and hydrogen
and oxygen. These gases are flammable and
explosive when mixed.
Agreement and understanding
The reader of this e-book assumes complete personal responsibility for
the use or misuse of the information contained in this book. The informa-
tion in this book does not conform to any known safety standards and it is
the reader s responsibility to adjust this material to conform to all applica-
ble safety standards after conferring with knowledgeable experts in regard
to the application of any of the material given in this book. The publisher
and author assume no liability for the use of the material in this book as it
is for informational purposes only. If these terms are not acceptable to
you, then don t read this book.
Clicking the box at right will open the e-book. Clicking this box
i
signifies that you have read and agree to these terms.
Build A Solar
Hydrogen
Fuel Cell System
by Phillip Hurley
copyright ©2004 Phillip Hurley
all rights reserved
illustrations and e-book design
copyright ©2004 Good Idea Creative Services
all rights reserved
Wheelock Mountain Publications
is an imprint of
Good Idea Creative Services
Wheelock VT USA
ii
How to use this e-book
Text links
Click on maroon colored text to go to a link within the e-book.
Click on blue colored text to go to an external link on the internet. The
link will automatically open your browser. You must be connected to the
internet to view the externally linked pages.
The TOC button will take you to the first page of the Table of Contents.
The left facing triangle will take you to the previous page.
The right facing triangle will take you to the next page.
iii
Table of Contents
Caution..............................................i Solar Panels (continued)
Basic solar panel components ........13
About this e-book..............................ii
Designing the ESPM ......................13
How to use this e-book ....................iii
Choosing PV cells ..........................15
Preface............................................vi
Testing solar cells ..........................17
Photovoltaic fuel cell systems
Tab and bus ribbon..........................21
The beginning of solar
hydrogen technology ....................1
Circular to square wire
conversion table ........................24
Primary components ........................2
Building the ESPM
Gas processing and storage ..............2
Tool list ..........................................27
Renewable energy sources and
hybrid generating systems............3
Materials list for two panels ............27
Solar hydrogen versus solar
Layout ............................................29
battery systems ..........................5
Tab ribbon and soldering ................32
Time limits for battery storage ..........6
Making strings of cells ....................39
Hybrid energy storage systems ..........6
Testing the connections ..................43
Solar Panels
Panel structure................................46
Types of solar cells ..........................8
Putting the cells into the frame ........53
Electrical characteristics
Power take-off box ..........................56
of solar cells................................9
Cover and final wiring......................57
Elements of photovoltaic
Designing and setting up a system
panel construction ....................10
of PV panels..............................60
iv
Task specific photovoltaic modules ..11
continued on the next page
Table of Contents
Electrolyzer Building the P41 Electrolyzer
(continued)
Electrolyzer basics ..........................74
Designing electrolyzers ................134
Electrodes ......................................77
Testing and measurement ..............136
The electrolyte ................................80
Safety ............................................83
Gas Processing System
About hydrogen ............................142
Water for the electrolyzer ................84
About oxygen ................................148
The P41 electrolyzer ......................86
Moisture and fuel cells ..................150
Building the P41 Electrolyzer
Gas processing components ..........151
Tool list ..........................................94
Building bubblers
Materials list ..................................95
Tool list ........................................160
Tank ..............................................97
Materials list for 2 bubblers ..........160
Gas exit port cap preparation ..........98
Other parts ..................................161
Separator preparation....................100
Assembly ......................................162
Positive electrode
assembly preparation ..............102
Installation in a system..................164
Negative electrode
Additions to the system
assembly preparation ..............109
Gas detection ..............................171
Installing the assemblies ..............116
Catalytic recombiners ....................172
Testing..........................................118
Building a recombiner....................174
Setting up the electrolyzer ............121
continued on the next page
v
Electrolyzer banks ........................129
Table of Contents
Gas Storage Planar Fuel Cell Stack (continued)
Preparing electrical connections ....217
Liquid phase and hydride storage ..179
MEA assembly ..............................219
Metal organic frameworks ..............180
Gas supply gasket ........................224
Low tech alternatives ....................181
PVC pressure plate ......................226
Double drum storage ....................182
Assembling the layers ..................227
Floating tank storage ....................184
Gas port installation ......................232
Calculating tank capacity ..............185
Testing the stack ..........................234
Adding pressure and safe storage..186
Trouble shooting............................237
Setup and check the system ..........187
Fuel cell power supplies ................240
Planar Fuel Cell Stack
Running the stacks........................245
Fuel cell basics ............................188
Resources
Types of fuel cells ........................189
Templates
Planar fuel cell stacks ..................192
More E-books
Building the
L79 planar fuel cell stack
Tool list ........................................194
Materials list ................................195
Electrodes/gas flow fields ..............198
Film transfer and etching ..............199
Routing the flow fields ..................208
vi
Plating the circuit ..........................214
Preface
A noteworthy conversation from  Mysterious Island by Jules
Verne, 1874:
It chanced one day that Spilett was led to say,  But now, my dear Cyrus,
all this industrial and commercial movement to which you predict a con-
tinual advance, does it not run the danger of being sooner or later com-
pletely stopped?
 Stopped! And by what?
 By the want of coal, which may justly be called the most precious of
minerals.
 Yes, the most precious indeed, replied the engineer,  and it would
seem that nature wished to prove that it was so by making the diamond,
which is simply pure carbon crystallized.
 You don t mean to say, captain, interrupted Pencroft,  that we burn dia-
monds in our stoves in the shape of coal?
 No, my friend, replied Harding.
 However, resumed Gideon Spilett,  you do not deny that some day the
coal will be entirely consumed?
vii
Preface
 Oh! The veins of coal are still considerable, and the hundred thousand
miners who annually extract from them a hundred millions of hundred-
weights have not nearly exhausted them.
 With the increasing consumption of coal, replied Gideon Spilett,  it can
be foreseen that the hundred thousand workmen will soon become two
hundred thousand, and that the rate of extraction will be doubled.
 Doubtless; but after the European mines, which will be soon worked
more thoroughly with new machines, the American and Australian mines
will for a long time yet provide for the consumption in trade.
 For how long a time? asked the reporter.
 For at least two hundred and fifty or three hundred years.
 That is reassuring for us, but a bad look-out for our great grandchil-
dren! observed Pencroft.
 They will discover something else, said Herbert.
 It is to be hoped so, answered Spilett,  for without coal there would be
no machinery, and without machinery there would be no railways, no
steamers, no manufactories, nothing of that which is indispensable to
viii
modern civilization!
Preface
 But what will they find? asked Pencroft.  Can you guess, captain?
 Nearly, my friend.
 And what will they burn instead of coal?
 Water, replied Harding.
 Water! cried Pencroft,  water as fuel for steamers and engines! Water
to heat water!
 Yes, but water decomposed into its primitive elements, replied Cyrus
Harding,  and decomposed doubtless, by electricity, which will then have
become a powerful and manageable force, for all great discoveries, by
some inexplicable laws, appear to agree and become complete at the
same time. Yes, my friends, I believe that water will one day be employed
as fuel, that hydrogen and oxygen which constitute it, used singly or
together, will furnish an inexhaustible source of heat and light, of an inten-
sity of which coal is not capable. Some day the coalrooms of steamers and
the tenders of locomotives will instead of coal, be stored with these two
condensed gases, which will burn in the furnaces with enormous calorific
power. There is, therefore, nothing to fear. As long as the earth is inhabit-
ix ed it will supply the wants of its inhabitants, and there will be no want of
Preface
either light or heat as long as the productions of the vegetable, mineral or
animal kingdoms do not fail us. I believe, then, that when the deposits of
coal are exhausted we shall heat and warm ourselves with water. Water
will be the coal of the future.
 I should like to see that, observed the sailor.
 You were born too soon, Pencroft, returned Neb, who only took part in
the discussion by these words.
x
Photovoltaic fuel cell systems
The beginning of solar hydrogen technology
The nineteenth century was an exciting time for electrical experimenta-
tion and discovery. Shortly after Alessandro Volta demonstrated the volta-
ic pile to the Royal Society of London in 1800, two experimenters, William
Nicholson and Sir Anthony Carlisle, discovered that hydrogen and oxygen
could be produced by passing an electric current through water. This was
the first demonstration of the principle of electrolysis.
In 1839 in Paris, nineteen year old experimenter Edmund Becquerel dis-
covered the photovoltaic effect when he found that certain materials would
produce electricity when exposed to light. In that same year William Grove
experimented with reversing the process of electrolysis and invented the
first gas battery or fuel cell. In the 21st century, these three discoveries
converge in photovoltaic fuel cell system technology.
1
Photovoltaic fuel cell systems
Primary components
The major components of a photovoltaic fuel cell system are:
1. A photovoltaic power source.
2. An electrolyzer gas source.
3. Fuel cells
The photovoltaic power source can be any commercial BSPM (Battery
Specific Photovoltaic Module), or can be an ESPM (Electrolyzer Specific
Photovoltaic Module). The electrolyzer can either be a PEM (Proton
Exchange Membrane) or an alkaline type.
Gas processing and storage
Secondary components of the system are a gas processing train and
gas storage vessel. The gas processing train can include different compo-
nents such as scrubbers, catalytic combiners, coalescers, filters and
flashback arrestors. Its purpose is to remove particulate contamination,
purify the gas stream, and remove excess water or other liquids.
The gas storage system may consist of low pressure storage tanks,
bags, tubes, medium pressure vessels and tanks, or high pressure tanks.
Other methods of hydrogen storage include liquid phase storage, metal
2
hydride, and nano-tube storage.
Photovoltaic fuel cell systems
Both the gas processing and gas storage system include valves and
tube or pipes; plus, depending on the system design, compressors,
pressure relief valves, gauges and other equipment specific to the method
of storage.
Renewable energy sources and hybrid generating systems
In addition to photovoltaics, other renewable energy sources such as
geothermal, tidal, hydro and wind power can be used to produce hydrogen
for fuel cells. Any of these sources can also be combined with photovolta-
ic panels in a hybrid system to power hydrogen production. All of these
renewable energy technologies, whether combined or stand alone, have
the virtue of no continuing costs of buying and transporting fuel, and thus
no dependence on regular supplies of fuels from outside and politically
unstable sources.
The choice of a renewable energy source to power a hydrogen fuel cell
system will depend on the geography, climate and other resources of the
location. I have chosen photovoltaics as the example power source for
hydrogen production because it is available to more people than either
geothermal, wind, or tidal. For a person without much technical skill, solar
power is easier to implement than wind power. It is also less costly up
3
front and not as dependent on location as the other RE power sources
Photovoltaic fuel cell systems
noted, so a PV hydrogen system can be portable. In comparison to other
renewable energy systems, photovoltaics also has these major advantages:
photovoltaic systems are, for the most part, easier to put up and maintain.
They have no moving parts to wear out or maintain, so a system is less like-
ly to break down; and they can be gradually enlarged to increase capacity,
whereas other systems require greater expense to make upgrades. This
does not mean that a photovoltaic system is ultimately less costly, but it
does mean that it should be possible to fit a PV system into a more modest
budget.
Of course there are exceptions to this. For instance, if you live in Iceland,
you would certainly want to explore geothermal energy as your power
source. Or, if you live in an area where there is a great deal of cloud cover,
wind power or hydropower might be more viable alternatives. Hybrid sys-
tems such as a combination of photovoltaics and wind (or any other combi-
nation) can be a good a choice for areas where the resources are present
but limited. If you are interested in wind power, many wind turbines are suit-
able for a stand alone power source for hydrogen production, or as a part
of a hybrid photovoltaic system.
4
Photovoltaic fuel cell systems
Solar hydrogen versus solar battery systems
To date, photovoltaics and battery technology have been practically
inseparable, combining quite well to convert energy and store it to meet
power needs.
However, there are problems with this combination. Spent batteries
become toxic waste. At peak power production times the excess energy
produced by the PV panels usually gets shunted and lost, because the
batteries have a limited charging capacity. For example, where I live in the
northern hemisphere there is plenty of sunshine during the summer
months to charge my batteries. But, since there is plenty of sunshine, I
don t need as much electricity at that time of year. There is less energy
demand, but an abundant supply of power. When the batteries are not
used as much, it takes less time to charge them, and the photovoltaic pan-
els simply bake in the sun, shunting energy to no purpose.
In the winter the situation is quite different. There is a lack of sunlight,
and I need more electricity. The batteries are always hungry for a charge
in the winter. This energy inequality can be addressed by adding a wind
turbine to a PV system. Fortunately, it is often windy at the times when
sunlight is lacking, so these power sources complement each other well.
5
Photovoltaic fuel cell systems
You could also just add more PV panels to your system, which might give
you enough power to get through the low sunlight periods, but it would
mean that more power is wasted in the times of peak sunlight.
Another possibility is to add a solar hydrogen fuel cell system, which
has the potential to operate without interruption because there are no
charging limits. During peak power production times, all the available
energy that the system can convert can be stored for future use, limited
only by the amount of hydrogen storage available.
Time limits for battery storage
Batteries have to be constantly cycled, that is, charged and discharged
within a certain amount of time, or they will degrade over time in their
energy storage capacity and become less efficient. Hydrogen, on the other
hand, can be stored indefinitely for use at a chosen time and season. The
carrying capacity of hydrogen as an energy storage medium cannot be
matched by present battery technology. This makes a solar hydrogen fuel
cell system a good addition to any photovoltaic system.
Hybrid energy storage systems
Some experimenters have combined battery storage and hydrogen stor-
6
age to see if there are any advantages to such a hybrid storage method.
Photovoltaic fuel cell systems
For instance, where BSPMs are set up for normal battery charging duty, a
diversion shunt is added so that when the batteries are fully charged, the
electricity from the solar panel is diverted to the electrolyzer to produce
hydrogen. Simple types of shunt devices that are not particularly energy
efficient when solely used for charging in a regular BSPM system are, in
this system, quite efficient.
A BSPM system can incorporate hydrogen production in a number of
different configurations. A basic hybrid energy storage system requires
several 12V panels with an approximate 10 amp output. These panels are
connected in parallel with a diversion controller. To operate with the high-
er voltage and to use it efficiently, the system would include a bank of
three electrolyzers connected in series (bipolar connection) that will use
four volts each. This will be discussed in more detail at the end of the
Electrolyzers chapter.
With a battery and gas storage system, you can have the advantages
that each technology provides, and have a base for making the transition
to a battery-free hydrogen system in the future.
7
Solar Panels
Types of solar cells
The basic energy producing unit of a photovoltaic power generating sys-
tem is the photovoltaic or solar cell. Solar cells can be made from a variety
of different materials, however the silicon solar cell is the most common,
well developed, and readily available, so it is what we will discuss here.
A silicon solar cell is a solid state semiconductor device that produces
DC (direct current) electricity when stimulated by photons. The three most
readily available types of silicon solar cells are the single crystal cell, the
poly crystal cell; and the vapor deposition type, often called amorphous or
thin film cell.
Of these three types of silicon cells, the single crystal cell is the most
efficient per exposed surface area in producing current. The poly crystalline
is the next most efficient, and the amorphous is the least efficient per
exposed surface area of the group. Single crystal cells are the most expen-
sive to manufacture, polycrystalline comes next as far as manufacturing
costs, and the amorphous type is the least expensive to make of the three.
This is, of course, reflected in how much the cells cost to the end buyer. In
practice poly crystal cells generate slightly less current than single crystal.
Thin film (amorphous cells) produce about half the current of a single crys-
tal cell for the same area.
8
Solar Panels
Although amorphous cells are the least costly alternative of the three
types, they are not a serious consideration for solar hydrogen production as
the cost of the added area for panel support structures and extra space
needed mitigates some of the savings derived from lower manufacturing and
purchasing costs.
Electrical characteristics of solar cells
Each cell, no matter what its size, will produce around .5 volts  some
more, some less, depending on the cell. So, if you took a 5" by 5" cell rated
at .5 volts and 4 amperes of current, and divided that cell into four separate
pieces, each smaller piece would still generate .5 volts. Although the voltage
remains the same for each piece, the current output for each would be only
about 1/ 4 of the original larger section, or about 1 ampere.
This is an important consideration. Using larger area cells in an ESPM
allows you to use fewer cells in each string, which saves time and work when
constructing panels, as there are fewer tab and bus connects to solder.
Other factors that affect the current output of a cell are the amount of sun-
light that stimulates the cell, and the temperature of the cell. The amount of
sunlight reaching a solar cell will vary from minute to minute due to particu-
lates in the air, moisture, and cloud cover. Daily variations occur due to the
9
changing position of the sun and the angle of the light striking the cells.
Solar Panels
These factors and the seasonal angle variations affect the current output
of each cell. Heat also affects cell output. When the temperature of the
cells rise, the output decreases.
Elements of photovoltaic panel construction
A single photovoltaic cell does not produce much voltage and the cur-
rent output is limited by its size. To augment either the voltage or the cur-
rent output, solar cells can be connected in either series or parallel.
Although it is not always the case, usually the sun-facing surface of the
solar cell is negative and the back side is positive.
When cells are connected in series to increase voltage, the negative
terminal on the face of one cell is connected with tab ribbon to the posi-
tive terminal of the next cell s back. This type of connection adds the volt-
age of each cell. For instance, for five cells that put out .5 volts apiece and
are connected in series, the leads at the top and bottom will give a read-
ing on a voltmeter of 2.5 volts.
For parallel, the faces of the cells are connected to each other and the
backs of the cells are connected to each other. With this arrangement, the
current output of each cell is added, but the voltage remains the same as
the output of one cell. Five cells connected in parallel would give a read-
10
ing of .5 volts at 20 amperes of current if the cells are 4 amps output.
Solar Panels
Connecting two or more cells to each other creates a string. A string is
a row of cells connected to each other in either series or parallel. Multiple
strings are then connected to each other in series or parallel to form the
whole of the solar panel or module.
Two or more solar panels connected together form an array. Arrays can
be connected in series, parallel or in series-parallel.
Task specific photovoltaic modules
All photovoltaic modules are classed as some form of TSPM (Task
Specific Photovoltaic Module). This means that every solar panel is
designed to perform a particular task.
Conventional solar panels are called BSPMs (Battery Specific
Photovoltaic Modules). Most panels sold commercially are of this type and
are designed solely to charge battery systems. These panels come in 12
volt, 24 volt and 48 volt configurations. Most have short circuit current rat-
ings of from 2 to 10 amperes.
ESPMs (Electrolyzer Specific Photovoltaic Modules) are uniquely
designed to match the power requirements of electrolyzer systems. This
makes them more energy efficient and economical for their intended use.
11
Solar Panels
Whereas BSPMs output higher voltages and lower currents, ESPMs out-
put higher current and lower voltages.
Although not always possible, the best engineering practice is to design
any power supply to match the specific appliance that will use that power,
so that as little energy as possible is wasted. Still, BSPMs can be used
efficiently for hydrogen production if they are used to power a bank of
electrolyzers rather than just one.
For instance, a two panel BSPM array connected in parallel will suffi-
ciently power three electrolyzers connected in series. For more current
you would add more panels in parallel to the array.
Another option is to insert a voltage divider, or DC to DC converter
between the BSPM and electrolyzer, which will give the correct current and
voltage for the electrolyzer. However, with this configuration there will be
some power loss. If that is not a concern for you, then it is a viable option.
12
Solar Panels
Basic components to build a solar panel
Solar cells
Bus ribbon wire
Tab ribbon wire
Sheet metal backing (or other rigid material)
Support frame
Cover (Plexiglass®)
Electrically insulating underlay (if you use a metal backing)
Screws, bolts
Power distribution box
Wire and terminal blocks
Schottky power diodes with heat sinks (optional)
Designing the ESPM
Since solar cells come in a variety of shapes and sizes, it is not possi-
ble to give exact dimensions for the particular panel you will build. Your
panel dimensions will depend on what size cells you are going to use, how
13
much power it will generate, and where it is going to be situated. These
Solar Panels
factors dictate the final
dimensions and the type
of materials used for your
particular panel.
I will show you how a
panel is designed and
constructed. This will
show you how to make
your own calculations
based on the materials
you have and your power
demands.
The first design consid-
eration is to determine
what the power needs of
the electrolyzer will be.
Front and back views of the finished panels
I decided on a photo-
voltaic power supply that
would provide somewhere between 10 to 20 amps, at 4 to 6 volts. This
would be sufficient to power my electrolyzer.
14
Solar Panels
Choosing PV cells
The next choice is the type of solar cells to use. Cells come in a wide
variety of shapes and sizes. The most frequently encountered shapes are
square, pseudo square
(square with angled cor-
ners) and round. I chose
a pseudo square shaped
single crystal cell that
was rated at 4 amps at
.55 volt per cell. These
cells were purchased as
cosmetically blemished
and off spec (sub stan-
dard) thus, they were
available at a reduced
price. Each cell is 5" in
length and width.
As stated earlier, there
are three basic types of
cells: the single crystal
15
cell, the polycrystalline
Solar Panels
type and the amorphous type cell. There are other types of cells such as
dye titanium dioxide, ribbon and others. For the most part, the single crys-
tal and polycrystalline are more readily available, and are well suited for
building an ESPM.
When shopping for cells, you can purchase new cells with no flaws or
new cells with flaws, either cosmetic or those that are off specification. Off
spec cells are cells that did not have the expected output when tested at
the factory. They are still good cells, but may not provide the current and
voltage that would make them suitable for a commercial panel. When con-
sidering off spec cells, keep in mind that the lowest output cell on a panel
is what the panel s final output will be. So, be sure that the lowest output
cell is within your acceptable limits.
Cosmetic flaws can be anything from chips off the sides or corners, dis-
coloration, or lack of reflective coating. Cells can be just cosmetically
flawed and putting out full output; or they can be cosmetically flawed and
off spec. For instance, a lack of reflective coating on parts of a cell (cos-
metic blemish) can reduce their output as they will reflect more and not
absorb as much light.
Higher current cells are larger and cost more per cell. However, larger
cells mean fewer tab and bus connections to make, and they reduce the
16
number of cells needed to produce a given amount of current.
Solar Panels
I have listed solar cell suppliers in the resources section, but you can
also call major solar cell manufacturers. Request their prices for cosmeti-
cally flawed and off spec cells, and find out what their minimum order is.
Popular internet auction sites can also be a possible source for cells,
although I recommend these with caution. Take your time, shop around,
compare prices, and consider customer service.
Testing solar cells
Each cell should be tested for voltage and current output before you sol-
der the tab ribbon to the cells. Even if you are using up-to-spec new cells,
test them before you solder. They might still be flawed or have been dam-
aged in transit and handling.
Solar cell manufacturers use an array of lights that closely mimic solar
output to test cells. This gives excellent results, but is expensive and not
really necessary.
The most low cost test option is simply to take the cells outdoors on a
very sunny day with a multimeter. The good news is that the sunlight test-
ing station is free. The bad news is that if it is not a perfectly clear day,
barely perceptible changes in light intensity occur within seconds due to a
17 variety of atmospheric changes. This can easily throw your readings off a
Solar Panels
bit. However, this is not a really major concern  ball parking it should be
sufficient. A fairly decent sunny day works for this technique.
With the multimeter, measure the open circuit voltage and short circuit
current for each cell. Write down the reading for each one. The cells do not
have to exactly match each other in voltage and current output. The point
is to match the cells so that all of the cells put out voltage and current at
or above what your target output is. The lowest single cell output will limit
all the others to its output level, so try to match them as closely as possi-
ble.
When I was testing the cells for this project, I waited for the first sunny
day to get out and test the cells. I tested and selected my cells, and con-
nected them together in a string, The next day, when I performed the sec-
ond test, I was surprised to find that my current readings were much high-
er than I had expected. What I thought was a very sunny day for the first
test was not as sunny as the day of the second test. I realized that there
must have been more particulates and or moisture in the air on the first
test day than on the second test day.
The time of the day, season of the year, and atmospheric particulates
and moisture will cause changes in the readings. Full output can be test-
ed best in the summer at high noon on a clear day. This doesn t mean that
18
Solar Panels
cells cannot be tested this way at other seasons of the year or times of the
day. Just take into account that output variations can occur because of
these factors. You will, of course, get lower output readings in the winter
than in the summer as the light has more atmosphere to traverse due to
the tilt of the earth. Other times than high noon or when the sun is at
zenith in your location will give you a lower reading. Despite all these vari-
ances, I find this to be an excellent method of testing cells.
If you test cells this way, make some sort of holder that can be tilted at
an angle, and that will hold the cell so that you can easily take readings.
You should be able to adjust the direction and angle of the device so that
it faces the sun as directly as possible. Be sure that when testing that you
do not shade part of the cell  it will affect the reading.
Laying a cell on a piece of copper clad circuit board is helpful for taking
readings. If you do use a copper clad circuit board, make sure the surface
is clean and that the back of the cell is touching the copper clad well. Lack
of good contact will give you a false or weak reading.
When you take your cell reading for either open circuit voltage or short
circuit current, touch one probe to the solder finger on the top side, and
the other probe to the copper clad board that is in contact with the solder
finger on the other side. To measure voltage with a multimeter you simply
19
Solar Panels
dial in  voltage, place the probes on either end, and note the voltage. To
measure current move the dial to current measurement and note the read-
ing. For this project you will need a multimeter that can measure at least
20 amps. You could also use a separate voltage meter and ammeter.
Another way to test cells is to take one cell and expose it to sunlight on
a very clear day, then test this same cell under an artificial light such as
an ELH projector lamp bulb which has similar characteristics to daylight;
and/or a daylight photoflood such as a BCA-B1 which has a daylight color
temperature of about 4800° Kelvin. Natural daylight has a Kelvin color
temperature that varies between 5000° to 6000° Kelvin.
Or, take a reading for one cell (this will be your control) under the arti-
ficial light at a specific distance. Then, take the other solar cells and see
if they are close in output to the control cell. This will give a relative com-
parison to a cell which you know has the output that you need.
Solar cell manufacturers test their cells using a xenon light source with
filters, under  AM1 conditions. An AM1 condition is when the sun is direct-
ly overhead on a very clear day. AM stands for air mass. The number indi-
cates the amount of air that sunlight has to travel through, and the result-
ing spectrum and intensity change due to the variations in air mass. The
change in spectrum is visible when you see more red during sunrise and
20
Solar Panels
sunset than at midday. The sunlight has more air mass to go through at
sunrise and sunset, so the Kelvin color temperature will be different at
those times than when the sun is directly overhead. In AM1 conditions, the
irradience affecting a surface is considered to be about 1000W/m2 and is
called one sun or full sun.
This irradience or insolation (a term that is a shortened version of
 incoming solar radiation ) figure is used to size photovoltaic systems. Of
course the amount of incoming solar radiation depends on a number of
factors such as cloud cover, moisture, and particulate concentration in the
atmosphere. Each geographic region has its particular climate character-
istics to be considered when calculating the number of panels required for
the photovoltaic fuel cell system.
Tab and bus ribbon
When you have decided what type of cell to use for your project, you can
move on to purchasing the tab and bus ribbon that will connect the cells and
strings of cells. Tab and bus ribbon are made from soft copper that is
pressed into a flat wire. This ribbon is tin coated to make it easier to solder.
Most tab and bus wire used for BSPMs is .003", .004", and .005" thick.
21
Solar Panels
For ESPMs, a .005 thick
tab ribbon is sufficient to
connect the individual cells
to the bus ribbon, but the bus
ribbon needs to be larger to
accommodate the larger cur-
rent carried. For the panels
described here, a 20 amp
capacity conductor is needed
for the bus wire. Most suppli-
ers of cells do not have bus
ribbon in the thicknesses
required. Enquire and see
what is available.
Tab and bus ribbon can be
purchased in the thicknesses
required and cut to any width from manufacturers, such as E. Jordan
Brookes Co. They cut the ribbon from larger rolls and can provide any
width and thickness of tinned wire desirable. Most tab ribbon widths sup-
1 1
plied by secondary cell dealers usually run about / 16" to / 8", and bus
3
ribbon is usually cut at / 16" widths. Measure the width of the solder fin-
22
gers on your cells to see if you can use a larger width cell connect tab
Solar Panels
wire. The more surface you cover, the better. These sizes are generally
adequate for most cells and solar panels. The 1/ 16" width is usually fine for
most cells. Ask your dealer what thickness the tab ribbon is.
For an ESPM as discussed in this book, use at least a .005 thick ribbon for
the cell connects. The bus ribbon needs to be able to carry the current of this
ESPM, for instance .010" thick, at 3/8" wide; or a .020" thick at 7/32" wide.
You can make your own tab and bus wire. Copper foil is available in
thicknesses from .002 to .021 from McMaster-Carr or another supplier. It
can be cut into thin strips to your particular specifications. You can also
use flat grounding braid for bus ribbon, as it is made for heavier current
carrying capacity. Grounding braid comes in a variety of thicknesses and
widths and is usually tinned. Make sure to use Tinnit® to tin the tab and
bus if you make your own  it also helps to minimize oxidation. You can use
round wire, either solid or stranded, for bus connections as well as cell
connects, if you desire, but the flat ribbon is probably easier to work with.
Square mils
It is easy to find the current capacity of round wire in a variety of pub-
lished tables for the photovoltaic enthusiast, but it is not so easy to find the
current capacity of ribbon wire. Not very many people engineer their own
23
PV panels.
Solar Panels
Although the current carrying capacity of a wire is based on factors other
than wire diameter (type of insulation, stranded or solid), wire gauge is a gen-
eral indicator, within limits, of the actual current allowable for a certain area
of a conductor.
If you are working with a conductor that is flat and of a certain thickness
and width, it is useful to be able to compare its current carrying capacity to
round wire conductors.
Wire gauge is usually indicated in what is called AWG or B&S. AWG stands for
American Wire Gauge, and B&S stands for Brown and Sharpe gauge. They are
one and the same. In the
table at right, wire gauges
are listed with their rated
amp carrying capacity and
their area in circular mils
and square mils.
Since you will be build-
ing a panel that delivers
20 amperes of short circuit
current, use the table to
see what size wire carries
24
that current adequately.
Solar Panels
Square mils/current capacity
One square mil equals .001". To find
Note that #14 gauge will carry 20
the current carrying capacity of square or
amperes, but you now need to know
rectangular shaped conductors such as
what size of flat ribbon wire will
bus ribbon, multiply the thickness in inch-
es times the width in inches. The result
carry that same amount of current.
will be the square mils of the conductor.
To get the current carrying
The table on the previous page gives the
capacity of the ribbon wire, simply carrying capacity based on square mils.
multiply the thickness in inches by
Voltage drop
the width in inches. This will give
To find the voltage drop for any length
the area in square mils. Take this
of wire run, multiply the resistance per
square mil figure and match it up
foot of the particular size conductor,
with one of the wire gauges listed
times the number of feet in the run, then
in the table on the preceding
multiply this times the current (amps) you
will be running through the conductor.
page. Then, multiply your square
The result is the voltage drop. For exam-
mil figure by .7854 to give the
ple, an ESPM with an output of 16 amps
area in circular mils. Whatever
at 4 volts and a wire run of 10' to the
gauge the ribbon is close to in cir-
electrolyzer through a #10 conductor, will
cular mils will tell you generally
give a voltage drop of 0.16v. This means
that at the terminals of the electrolyzer
what the current carrying capacity
the system will be able to deliver 3.8v.
of the bus ribbon is. Note that this
table is only correct for copper
R x Ft. x Amps (resistance x length of
run in feet x amps) = voltage drop.
25
conductors.
Solar Panels
7
As an example, consider a bus ribbon .020" thick and / 32" wide.
7
Convert / 32" into its decimal equivalent (.21875) and move the decimal
place to 218.75 because 1 mil = .001". Then multiply 218.75 times 20,
which gives you 4375 square mils. Either check the table or convert these
square mils to circular mils by dividing 4375 by .7854, which is 5570 cir-
cular mils. As you can see from looking at the conversion table, this figure
falls between a 14 and 12 gauge, and any wire size within this range
would be sufficient to carry 20 amps.
26
Solar Panels
Building the ESPM
Tool list
All available at local hardware or electronics stores.
Screwdriver
Hacksaw, and other types of saws to cut metal and other items
Ruler, T square, other measuring devices as needed
Exacto knife, razor knife, other cutting instruments
Caulking gun
Soldering iron, 60 watt, with screwdriver-type tip
Paint brush to spread caulk
Drill, either hand or drill press
Multimeter that can measure current
Materials list for two panels
All should be available at local hardware or electronics stores,
unless otherwise noted.
40 PV cells, Plastecs or other (see Supplier URL list, Resources).
Tab and bus ribbon, E. Jordan Brooks or other (see Supplier URL list)
Fiberboard 28"x 30"
Clear silicone rubber caulking
27
1
20 Stand offs to hold plexiglass in place, /4  wide, rubber or plastic
Solar panels
Materials list for two panels (continued)
1
Silicone solid rubber, /32" thick, 12"x 36 , 30A durometer, McMaster-
Carr, part #8622K31
Tinnit® electroless bright tin plate, All Electronics, part #ER-18
16 Screws, 10-24 11/2  long, and nuts to fit (both stainless steel)
1
2 Aluminum sheets about /16" thick: 28"x 30" Local hardware store,
local metal supplier, sheet metal shop, or McMaster-Carr
Wire-red and black zip cord. All Electronics, automotive store. All
Electronics part #WRB-10 (10ga.) or WRB-12 (12ga.)
2 Wire connectors, crimp on type
2 Junction boxes, plastic case, 4.7"x 2.6"x 1.55" All Electronics, part
#1591-CSBK
2 Terminal strips, 2 position, dual row, All Electronics, part #TS-250 or
equivalent.
1 Terminal strip, 4 position, dual row, All Electronics part #TS 6034 or
equivalent.
Electrical tape, liquid electrical tape, liquid rubber or shrink tube
2 Sheets nylon screen 28"x 30",
3
2 Sheets Plexiglass® about /32" thick, 28"x 30", McMaster-Carr or
hardware store
28
Aluminum bar stock .25" thick, by 1" wide, hardware store, local metal
supplier, sheet metal shop, McMaster-Carr
Solar Panels
Materials list for two panels (continued)
Solder  can be 2-4% silver solder or regular 60-40 depending on what
works for your particular PV cells. 2-4% silver is recommended
although it may not be necessary. Do not use acid flux!! Use rosin
Flux pen, 2 pens, or as needed. Do not use acid flux. HMC Electronics,
#186FP Mildly activated rosin, type RMA
Crazy glue or epoxy
2 Schottky diodes, 45prv,75 amp, Surplus Sales Of Nebraska, part
#SDI-USD5096F (optional). 2 Heat sinks for diodes, Surplus Sales
Of Nebraska, #HSK-HEATCNK90 (optional). Distribution box to
house diodes and heat sink (optional).
Grommets for junction box and distribution box (optional)
Building the ESPM
Two ESPMs were constructed for this project. The design was based on
materials that were on hand, so the finished products are not optimal, but
were relatively inexpensive. For each panel back, I used a 28" x 30" sheet
of rigid aluminum metal that was a little less than 1/ 16" thick. I purchased
a piece of fiberboard and cut it to 28" x 30" (the same dimensions as
planned for the finished panels) so that I could use it as a  peel, a sol-
dering platform, and layout grid for the cells, tab and bus ribbon positions
29
for the panels. I used 4 pieces of 1/ 4 x 1" aluminum bar stock for a
The over-all
layout for the
solar panels,
showing place-
ment of rubber
spacers
30
Solar Panels
frame/support and as a spacer between the back of the panel and the
cover. For the covers I used 28" x 30" x 3/ 32" Plexiglass® sheets.
With a pencil I outlined an inch border on all four sides of the fiberboard.
This one inch border marks the space taken up by the aluminum side
bars/frame of the panel.
The particular cells that I Detail of spacing for cells,
bus and tab wire.
had were 5" in width and
height. I laid out the cells on
the fiberboard to see how
many I could fit in the 26" X
28" space with enough
space left for tab and bus
ribbon connects as well as
enough space to stay com-
fortably away from the metal
edges. With these cells I
could just fit twenty cells (4
strings of 5 cells each) with
sufficient space for tab and
bus ribbon connects, and
31
power take off leads.
Solar Panels
After calculating what looked like a pleasing and practical configuration, I
outlined every part that would be laid on the board. This outline was my
guide for soldering and connecting the cells and strings of cells.
Cutting the tab ribbon
The individual cells have two tab finger lines on each side of the cell, so four
tab ribbons had to be cut for each cell. The tab ribbon runs the total length of
the cell, so the length for each tab would have to be 5", plus an extra 1/4" for
space between the cell and the bus ribbon connect, plus the width of the bus
ribbon that the tab is connecting on to. So, I needed to cut 160 pieces of
57/16" long tab ribbon for the 40 cells that would be in the two panels.
Crimping the tabs
For these panels, I did not crimp the tabs, however, you should crimp the
tabs so that your connections can expand and contract. Panels are usually
exposed to environments with wide temperature fluctuations, from very hot to
very cold. Crimping permits slight movement and gives flexibility that helps to
ensure that the tabs stay connected to the cells and bus ribbon. Temperature
extremes cause expansion and contraction that can break the connections
over a period of time.
32
Solar Panels
If you add a crimp to the cell tabs,
1 7
add another / 4" to the 5 / 16"
1
length mentioned above. The / 4"
will allow a 1/ 8" high crimp. It can be
a little less if you wish, but the crest
or raised part of the crimp should be
centered so that it falls mid point
between the cell and the bus ribbon.
Preparing the soldering iron
to tin the tabs
Tinning the tabs
Tinning is a simple process in
which you touch the tip of the sol-
dering iron with the solder wire, and
cover it very liberally with solder.
The solder melts on the tip and
forms a glob which can then be
deposited on the surface of the tab
ribbon as you run the iron up and
Tinning the
down the length of it. Coat the rib-
tab ribbons
bon wire completely and generously
for the length that is needed.
33
Solar Panels
The purpose of tinning is to
make a solder base from which to
form a joint when the ribbon is
attached to the cell fingers or to
tab or bus ribbon. Although the tab
and bus wire comes already coat-
ed with a thin layer of tin, it is still
necessary to apply more tinning
on the areas to be joined.
Practice tinning a few pieces of
Soldering tools
tab or bus ribbon to get the knack
of it. If you miss areas that should be tinned, the joints might not be as solid
and strong as you will want them to be. Try to coat as evenly as possible by
running the iron down the length of the ribbon. If you have not finished the
entire length and the solder is thinning out too much, put more solder on the
tip and continue until the surface is fully and evenly coated.
Each tab was tinned from one end for a five inch length on one side (to be
connected to the cell); then, on the same side, but on the opposite end, a
3
/ 16" length was tinned (to be connected to the bus ribbon).
34
Solar Panels
This can be done according to how you want
the finished product to look. If you want the bus
ribbon to cover all the tab ends when you look at
them from the front of the cells, then tin the
3
/ 16 on the same side for half of them (for one
3
panel that would be forty) and / 16 on the
opposite side for the next forty (see illustration
at right). If you do this, when the bus ribbon is
laid down to be soldered to the tabs, all the tab
ends will be under the bus ribbon. It gives a
neater finished look to the panel, but is not nec-
Tinning crimped tabs
essary and will not affect performance.
for a neat finish.
If you have crimped the tab ribbon, be sure not
to get solder on or in the crimp as this will make it rigid and inflexible. This
would negate the purpose of the crimp, which is to add flexibility to expand
and contract with temperature changes.
Cells from different manufacturers have different kinds of solder fingers
on the faces and backs of the cell. Some of the older types of cells that
you may run across have the back of the cell completely coated with sol-
der, and the fronts have thick solder fingers. Newer types of cells usually
35
Solar Panels
have the metal fingers put on by a vapor deposition technique, or they are
silk screened on. The conductive metal in these deposited or screened fin-
gers can be nickel, silver or other conductive alloy. Because of the various
methods of deposition and the nature of the conductive metal used, be
sure the soldering materials you use is compatible with the solder fingers
on your PV cells.
Generally, use a solder with a 2% to 4% silver content. A 96/4 (96%
tin, 4% silver) silver solder is generally available at most electronic
stores. The melting point of 96/4 is about 460°F. However, I found that I
could use regular 60/40 (60% tin, 40% lead) on the cells I had. I applied
solder to the tabs liberally during tinning, and it soldered quite well to
the cell fingers. 60/40 is less expensive and if it works well on your cells,
use it. The melting point of 60/40 is about 430°F, which is a little lower
than the tin/silver alloy.
When you purchase cells, ask which type of solder will work the best. If
the recommended solder is a silver type solder, try it and see how well it
works, and then try some regular 60/40 and compare. Tabs will tend to pull
off easily after soldering if you do not have enough solder tinned to the
surface of the ribbon to make the joint.
36
Solar Panels
Fluxing the cell fingers
The secret to a good bond, especially
when working with solar cells, is using the
right flux and using it correctly. Flux cleans
the metal surface and reduces surface ten-
sion between the solder and the metal it
will adhere to. This is important to make a
good electrical contact. Do not bother with
paste or liquid flux. Use the flux pens indi-
Applying flux to
the cell fingers
cated in the materials section. The pen will
put the right amount of flux down and do it
very quickly and easily when working with cell fingers. It does not waste
flux, and one pen was enough for this two panel project. The tip of the pen
is perfect for cell tabbing. Flux pens have a felt tip which is spring loaded
with flux when you push the tip against a surface. When working with solar
cells, do not load the felt tip while the tip is on the cell. Push the tip
against a surface other than the cell. This will load the tip with liquid flux.
Then, take the tip and run it down the cell finger. It will cover the entire fin-
ger with flux. Do not press anything against the cells  they are fragile and
liable to crack. Using a flux pen rather than other types of applicators is
37
very fast, easy and efficient.
Solar Panels
Apply the flux to one tab finger at a time;
that is, flux a finger and then immediately
solder a tab to that finger, Then, move on to
the next finger, flux, and tab, and so on.
Soldering the tab ribbon to the fingers is
easy. Position the tab on top of the finger
and hold one end with a stick or other
object to keep it steady and aligned while
Soldering tab ribbon to the
cell front (above) and cell back
you solder. Then move the iron down the
(below).
length of the tab. You will see the solder
melting as you move. Do not leave the iron
stationary in one place too long, as the heat
from the iron can damage the cell. When
moving the iron along the tab ribbon, don t
move too slowly or too fast, but just the
right speed to melt the solder along the
length as you proceed. Practice on some
scrap cells to get a feel for the timing.
38
Solar Panels
Making a parallel connected string
After all of the cells needed are tabbed, connect them into strings, accord-
ing to your design. My panels called for four strings of five cells each.
Position the tabbed cells within the cell outlines drawn on the fiberboard,
and tape each cell by the tab ribbon to the fiberboard so that the cells will
not move. Do not put tape over the surface of the cells as this can leave a
gummy residue, and when you take the tape off, it can crack the cell.
Next cut eight pieces of 26" long bus ribbon and lay them out in the bus
ribbon position drawn on the fiberboard, over the tabs which extend from the
cells. Tin two pieces of bus ribbon at the points where the tab ribbons will
connect with them, and then tape the bus ribbons in their positions on the
board on either side of the cells. Although the tape is not necessary, its
39
String of PV cells connected in parallel
Solar Panels
helpful to keep everything steady
while you are soldering. Tin the iron
lightly, and apply the iron to each
tab solder point  eight soldering
points on each side of each bus.
This makes a .5 volt 20 amp string
with a negative bus on one side and
Solder the tabs to the bus wire
a positive bus on the other side.
to connect the cells in a string. Note
Test the short circuit current under
that this is a positive bus wire, con-
necting to the PV cell backs.
sun light by putting the multimeter
probes on the ends of the negative
and positive bus leads, then test
Trim the tab ends that extend
the voltage. If you are not getting
beyond the bus wire. Note
that this is a negative bus wire,
the reading you should be getting,
connected to the PV cell faces.
check the solder joints and resolder
if necessary. Solder the other
strings in the same manner on the
guidelines of the fiberboard and you
will then be ready to connect each
of these tested parallel connected
strings of cells in series to each
40
other.
Solar Panels
Connecting the strings
in series
With the strings laid out in the final
positions they will occupy in the
panel, connect the strings with bus
ribbon. For my panel, I cut 18 short
pieces of bus ribbon the length need-
ed to connect the string buses that
lie next to each other (see illustration
at right). Tin 3/ 16" on each end of the
short pieces of bus ribbon.
These strings will be connected to
Panel layout showing
electrical connections
each other in series, which means
that the positive bottom side of each
string is connected to the negative top side of the next string, and so on.
The bus bar on the left side of the first string on the far left will be neg-
ative. The bus bar on the right side of this first string will be positive.
This positive bus is connected to the negative bus of the string next to it
and so on. Make sure that when the strings are laid out on the board
they are all aligned the same way, for instance, the left bus of each
41
string is negative, and the right bus of each string is positive. With this
Solar Panels
42
Soldering the strings together in series.
Solar Panels
particular arrangement, the negative bus ribbon that will connect to the
take off will be on the left edge of the panel, and the positive bus ribbon
that will connect to the take-off will be on the right edge of the panel.
Be sure that the layout is correct and that everything is taped down.
Soldering the connections
Solder the buses together with the small pieces of bus ribbon intercon-
nects. Face the tinned side down on the bus ribbon and apply the iron on
each end while securing the other
end with a stick or other tool so that
it doesn t move when you solder
(see illustration at right). When the
string buses are all connected, test
the voltage and current again.
Test outdoors
With everything still taped to the
Connect the strings with
small pieces of bus wire.
fiberboard, take the layout outdoors
and tilt the board toward the sun at
43
Solar Panels
an angle to give it maximum sun. Take the open
circuit voltage reading by placing the one multime-
ter probe on the far left bus ribbon, and the other
probe on the far right bus ribbon at the top of the
strings. This is where the power take off bus will
connect with the strings. This reading indicates the
total voltage, which should be the sum of .5 volts
Trim any excess
per string. In the case of the panels that I built, this
bus wire when
was around 2.0 volts.
the strings are
all connected.
Next, measure the current output by changing
the setting on the multimeter to test current. Again
place one probe on the far left bus ribbon and the other probe on the far
right bus ribbon. This gives the short circuit current reading. The panel I
built gave a reading of around 20 amps. You may get a lower reading
depending on the time of day and season. etc.; but the reading should be
somewhere from 16 to 20 amps. As mentioned before, it is best to wait for
a clear and bright sunny day and go out about noon when the sun is high-
est in the sky. Make sure the fiberboard is tilted to take full advantage of the
sunlight. This will give your maximum reading.
If you are not getting the voltage or current readings that you should,
44
visually inspect the cells and the connections. If you cannot see a flaw, try
Solar Panels
probing with the multimeter and take readings on the strings and cells at
different locations to try to find the problem. If one of the cells is cracked,
replace it by applying the soldering iron to the joints to desolder and
remove it from the string, then replace it with another cell.
Usually you will not encounter any problems in the final test, if you test
as you proceed with assembly. The first test should be on each cell indi-
vidually to check general output. The second test occurs when you finish
each string of cells to make sure that the string's output is what it should
be. The third test is to take readings when all the strings are connected to
each other. This tells you the final output value of the panel.
Handle cells and cell assemblies gently
During the construction process be sure to handle and move the cells
carefully. I usually have several pieces of fiberboard on hand so that I can
peel/slide the cells from one board to the other very carefully and quickly
when I need to make space while I am soldering. If you pull or stress the
ends of the cell tab connects, you can inadvertently pull the tabs off the
cell. The keyword here is gentle! If one of the soldering joints does come
loose, resolder as needed.
45
Solar Panels
Building the panel structure
When the strings of cells have all been connected, you are ready to
assemble the rest of the panel.
For the ESPMs I made, I used precut 28"x 30" sheet aluminum that was
about 1/ 16" of an thick. For the side bar/framing material I used 1" by 1/ 4"
aluminum bar stock cut into four 28" pieces for each panel. For the covers
3
I used / 32" thick Plexiglass® that was cut at the hardware store to 28"x
30". When you purchase a cut Plexiglass® sheet, it usually has a thin film
on either side to protect the surface from scratches. Leave this film on
while you drill and until you are ready to finally assemble the panel. It will
protect the Plexiglass® from scratches during construction.
With all components cut to size, the next step is to drill the holes for the
screws in the side bars, panel back, and plastic cover; and two holes for
the take off leads on the aluminum panel back.
For these particular panels I wanted to have four screws along each
edge for a total of 16 screws to hold the structure together. More screws
can be used if you wish. To drill the side holes, c-clamp the three compo-
nents together, with the aluminum sheet on the bottom, the Plexiglass®
cover on top of that, and the side bars on the very top.
46
Solar Panels
Align all the pieces to each other and c-
clamp all four edges so that nothing moves
while drilling the holes. Then, drill the holes.
Drilling the holes in this manner assures
that the screw holes will line up as they
should. You can drill each piece separately,
but be sure that your measurements and
drilling technique are very accurate.
Clamp all the structural
Before you unclamp the pieces, mark
components together, then
each piece so that you will know exactly
drill the holes for the screws.
which piece goes where, when you are
ready for final assembly. This is very impor-
tant, as minor variations can cause assembly problems if the holes don t
match exactly.
Power take off box
After drilling the screw holes, measure, mark and drill the power take off
holes. One way to do this is to construct the power take off box first. Any
size project box from any electronic supplier or store can be used.
To prepare the box, first make sure the surfaces to be glued are flat. In
47
other words, remove any mold marks from the outside back of the box, so
Solar Panels
that it will sit flat against the
aluminum panel back. Sand
the surface of the box to be
glued to give added grip.
Then remove any mold
marks inside the box on the
surface where the connector
block will be glued. Place the
connector block inside the box
to determine the position for
The power take-off box, parts (above)
the take off holes in the plastic
and assembled (below)
box. Also, figure where the
holes for the lead outlets to the
electrolyzer will be. To choose
an appropriate hole size, be
sure to consider the width of
the bus ribbon including shrink
tube insulation. Outlet holes
going to the electrolyzer
should be sized for the wire
48
used, plus additional space if
Solar Panels
you use rubber grommets in the holes. Rubber grommets can be slipped
into holes to protect the wires from fraying and to partially seal the holes
from moisture. Mark the take-off and the lead outlet holes and drill them.
Sand the inside surface of the box where the connector block will be
glued, and also sand the back of the box where it will come into contact
with the back panel surface. After sanding, clean the surfaces of all grit
particles or other foreign material. Then, glue the connector block into the
box with an instant glue or epoxy.
When the outlet box has dried, position it on the back of the panel where
you want to put the take off holes. Mark the take off holes through the box
onto the back of the panel and mark the outline of the box on the back of
the panel. Drill the holes in the panel back for the power take off ribbon.
Prepare the backing surface to lay screen
The next step is to sand the face of the panel back to lay the screen,
which is mounted with silicone caulk. The sanding gives a rough surface
that helps the silicone adhere. The bottoms of the bar stock (side bars)
that will abut the panel surface should also be sanded for this reason.
After drilling and sanding, wash all metal parts down so that no metal
powder or flakes are left on the surface. Tiny metal particles from drilling
49
Solar Panels
and sanding can short out the panel and cause problems later. After wash-
ing, be sure the parts are dry before applying the silicone.
Glue the side bars to the panel back
(See the photos on the next page.) Run several beads of silicone caulk
along the face of the panel backing where the side bars will be placed.
Insert the screws up through the panel back, then slide the bars on to the
screws. Put the nuts on to hold the bars securely. This will hold the bars
in place while the silicone is drying. Then, run your finger along the inside
edge to smooth out any excess silicone that is squeezed out from under
the bars. Put another bead along the inside edge where the bar meets the
back and smooth that out with a finger. This will create a moisture seal.
Smooth the outside edge and then apply more silicone and rub along the
edge with a finger to smooth it out. You can wear latex gloves for this. Give
the silicone about 24 to 36 hours to dry, then take off the nuts and remove
the screws.
Apply the screen
Cut a piece of fiberglass screen to fit within the frame/side bars. For
these panels with overall dimensions of 28"x 30", I subtracted the width of
the side bars (1" each) from the overall dimensions, so I cut the screen to
50
Solar Panels
51
Glue the side bars to the panel backing
Solar Panels
26"x 28". The screen
insulates the PV cells
from the metal back.
Cover the surface
of the face of the
panel backing with
silicone caulk, in an
even layer with no
missed spots. The
best way to do this is
with your hands.
Squirt the silicone
from the caulk gun on
to the panel back and
them smooth it with
your hands (see pho-
tos at right). Work
quickly, because the
layer of silicone is
thin and will dry fast.
Apply a layer of silicone to lay the screen.
52
Lay the screen in the
Solar Panels
silicone on the face of the panel backing. It
should fit in perfectly. If not, trim any excess
with a razor knife. Set it aside and let it dry
for 24 to 36 hours.
Lay the screen on the silicone.
Putting the cells into
the frame
Peel/slide the cells into the
frame. Use the fiberboard guide
to gently slide the cells onto the
frame, carefully aligning the
cells to their final placement.
Dab silicone along and under
the bus ribbon of the connected
strings (see photo, next page)
Do this well, as this silicone is
53
Solar Panels
what holds the cells in place when the panel is
upright. Be careful not to drop silicone on the
cell surfaces.
Place 10 rubber spacers in place with silicone
adhesive (see the illustration on page 27).
These spacers hold the Plexiglass® away from
the cell surfaces so that the cells do not break,
and they support the Plexiglass® panel cover.
They can also be made from heat resistant
plastics. In placing the spacers, be sure that
Dab silicone under
they do not cast a shadow onto the cells when
the bus wire to hold the
the sun is at an angle to the panels.
strings of cells in place
Next, cut two holes in the screen for the
power takeoff holes. Allow the silicone holding the cells to dry for 24
hours.
Take two pieces of bus ribbon (mine were about 16" long) and solder
one to the left side at the top of the bus, and connect the other bus rib-
bon to the right side top. Put a piece of cardboard under the bus ribbons
54
Solar Panels
that you are soldering so
that the heat from the sol-
dering iron does not melt
the nylon screen.
Insert the power take off
leads through the take off
holes in the panel and
Solder the connection to
notice where they touch the
the power take of leads.
panel as they go through.
Put shrink tubing on the take off
leads so that the leads are insulat-
ed from the metal backing of the
panel. To do this, slip a piece of
shrink tubing over both bus wire
leads, hold it in its final position,
and apply heat to shrink the tubes.
Push the bus wire through the
holes again and apply silicone
under the take off bus wire to hold it
Use heat shrink tubing
to the frame/screen surface. Then,
55 to insulate the leads.
Solar Panels
dab a little silicone around the edge of the take
off holes, both front and back, for insulation.
Bend the bus wire toward the panel back to
hold it so that it doesn t flop around while it is
drying. Let it dry for 24 to 36 hours before mov-
ing the panel to an upright position.
Attaching the power take off box
With either epoxy or instant glue,
cover the back of the box and the sur-
face of the aluminum back that the box
will adhere to. Push the bus wire
through the holes in the panel back
and box. Position and press the box
into place. If using epoxy, let it set
before going on to the next operation.
Apply silicone all around the edges of
the box that meet with the panel back.
This will help to seal against moisture. Let
Pull the bus wire through the holes
it dry.
in the panel back and terminal box.
56
Solar Panels
Attach the bus leads to the screw terminals If the bus wire is wide, trim
it so that it just slides into and under the screw heads in the terminal. Cut
off the excess bus wire. Another option for wide bus wire is to drill a small
hole in the bus wire for the screw in the terminal connection.
Placing the cover
1
Adhesive backed / 32 silicone rubber is
used to make a seal between the side
bars/frame and the Plexiglass®. This materi-
al comes with a paper backing, which makes
it easier to handle, but once the paper back-
ing is removed, the rubber shrinks a little in
length. Leave the paper backing on initially to
cut the rubber.
The 3'x 1' rubber sheet should be cut in
strips the same width as the side bars. To fig-
Silicone rubber strips
are used to make
ure the length to cut, two strips should be the
a gasket between
height of your panel, plus an inch for shrink-
the side bars and the
age. The other two strips should be the width
Plexiglass® cover
of your panel, minus the width of two of your
57
Solar Panels
side bars, then add an inch for shrink-
age. This way, at the panel's corners
the seams between the pieces of rub-
ber are staggered with the seams
between the side bars, which makes a
Three layers: aluminum side
better seal (see illustration at right). For
bars, silicone gaskets,
Plexiglass®. Note that the
each of my panels I cut two 31" strips
gasket joint is staggered
and two 27" strips, all 1" wide.
with the aluminum
side bar joints.
Remove the paper backing from the
rubber, and let it shrink. If you haven't
already done so, remove the screws from the side bars. Lay the longer
vertical strips of rubber on the side bars/ frame, so that each strip covers
the full vertical length of the panel, then trim off the excess rubber. Lay in
the shorter horizontal strips, and trim them so that they fit snugly against
the vertical strips. At the edges where the strips meet, apply silicone to
improve sealing.
Remove the protective plastic from the Plexiglass®, and lay the
Plexiglass® over the panel, lining up the screw holes. Before inserting
each connecting screw, dab a bit of silicone along the thread area and
under the head of the screw. Also put a dab of silicone in the screw hole
58
Solar Panels
itself. Insert the screws down through the top.
Before you put the nuts on to secure the panel,
dab silicone where the screw comes out of the
back of the panel, and then put the nut on.
When the nut is tightened on each screw, dab
some more silicone over the nut. This helps to
make the panel water tight.
Installing the
Finally, rub silicone along the edges of the
Plexiglass® cover.
panel to ensure a final water tight seal.
Finish wiring the panel
Connect the takeoff wire (the wire that is going to the electrolyzer)
through the holes in the junction box. If you use grommets, be sure to put
those in first. Slip the wire through the box and either make a knot in each
wire or use a wire hold of some sort to prevent the wire from being yanked
from the terminal screw, from outside the box. Connect the red lead to the
positive terminal and the black lead to the negative terminal.
59
Solar Panels
Designing and setting up your system
Reducing circuit loss
Reducing circuit loss to a minimum is an important goal when designing
a system. The three most important things to do to reduce circuit loss are:
1. Use the shor test wire run possible.
2. Use the largest diameter wire possible.
3. Reduce the use of diodes as much as possible, or use low
voltage drop Schottky diodes.
The rating of the diode should always be more than the open circuit volt-
age and short circuit current produced by the individual panel or array
used. Diodes also need aluminum heat sinks to dissipate the heat gener-
ated within the diode by the flow of current.
Connectors and switches should be of decent quality and all compo-
nents should be sized correctly for current carrying capacity. The
less resistance you introduce into a circuit, the less voltage drop you
will experience.
Corrosion will create resistance in the circuit, so, it is important to
house the components, switches, diodes, connectors, etc. in weatherproof
60
Solar Panels
housings if the compo-
nents are to be out of
doors for any length of
time. A very slight oxida-
tion on contacts can
develop great resistance
and reduce power output
drastically. Connectors
and switches need to be
protected from moisture
and rain to prevent short
circuits, so make sure the
housing for them is wet
proof. Most electronic/
electrical suppliers have
Connecting block configuration
boxes with rubber seals
for two ESPMs connected in series.
that are made to be
weatherproof. You can also
seal boxes with silicone caulk. Connectors can be covered with liquid electri-
cal tape or a dip type rubber compound to insulate them.
61
62
Configuration options for connecting PV panels to electrolyzers
Solar Panels
Grounding
ESPM panels should be grounded for long term service to protect
against lightning strikes and electrostatic charge buildup. Even distant
lightning strikes can induce EMF in the panels and cause problems. Wind,
pollen, dust, and snow blowing across the panels can also cause electro-
static charge buildup. Grounding braid can be connected to the panels on
one of the long screws that jut out on the back of the panel, with the other
end of the braid securely attached to a ground rod. To add circuit ground-
ing, use a lightening arrestor for added protection.
Diodes
With the simple system described in this book, no blocking diode is
needed because the  dark current (night time reverse flow) is negligible
from this electrolyzer and will not harm the panels. For a large bank of
electrolyzers, consider putting in a blocking diode. Blocking diodes are
used in BSPM systems primarily to prevent back flow of current from the bat-
tery into the panel at night, which depletes the batteries. However, there is
some question about the usefulness of blocking diodes for less than 24 volt
BSPM systems because charging current is lost through the diode during
battery charging anyway.
63
Solar Panels
A solar hydrogen system that does not have
a hybrid storage system (both batteries and
gas storage) or doesn't use BSPMs exclusive-
ly, will have no loss of stored energy back into
the PV panels. Diodes can be used, but, for
this system they are not necessary.
Blocking diode
Switches
with heat sink
An on/off switch can be useful. It can be as
simple as a weatherproofed toggle switch
rated above the amperage of the panel or
array. Such a switch in a solar hydrogen sys-
tem should be placed close to the panel and
not near the electrolyzer unless it is explosive
rated switch. A simple connector box which
contains the contacts and/or diodes can also
house the switch.
A low cost wireless switch can be installed in
the box to turn the electrolyzer on or off at a
distance. A wireless float switch could be used
64
Solar Panels
to turn off the electrolyzer when the gas tanks are filled. The details of such a
switch will depend on the use and/or storage methods for the gases involved.
Data loggers
Data loggers can be integrated into the system to record volts and amps.
The data can be downloaded to a computer from the field, or be directly con-
nected, either hardwired or wireless, to transmit data to the computer. This
will give a very accurate record of the system's performance on a minute to
minute basis, and give you the capability to correlate panel output to gas
production, daily insolation, etc.
Simple voltmeters and ammeters can be added to the circuit for quick
visual indicators.
Fuses, connectors and cables
If you are interested in a hybrid storage system with a battery, you can use
a blocking diode and you should definitely fuse the battery, or use MCBs
(miniature circuit breakers) to prevent possible problems from a short.
Battery connections should be very secure. Do not use alligator clips as
they have a tendency to be easily knocked off and spark. Always use weath-
er proof boxes for connections and switches. If you are going to run cable
65
underground and or have it exposed to sunlight outdoors, make sure it is
Solar Panels
rated for that purpose, that is, type TC or UF. This cable can be wired to
the power box where you have the switch.
For panel to panel connections, a #12 or #10 red and black zip cord is
fine, and, depending on your setting, may be all that is needed for your
connections. If you do not have underground runs or extensive (more than
10 ) outside runs, or you do not expect to keep the system up for exten-
sive periods of time, then the zip cord is probably all you need. The
integrity of your wiring should be checked regularly. Outdoor wiring can be
coated for UV protection. Ultimately, refer to any codes for your local area
for the final word on wiring requirements.
Positioning solar panels
To maximize the output of a solar array, angle the panels according to
your latitude and to the season. To find your latitude, look at a paper map
or atlas, or access the US geological survey map information URL via the
Resources page at the end of this book. There are also many other sites
on the internet that give this information for wherever you live in the world.
On the USGS site, for instance, all you have to do is put in the name of
your town, state and country.
Where I live in Vermont in the northern hemisphere, during winter the
66
sun is low on the horizon as it transits east to west. During the summer
Solar Panels
the sun is much higher and transits overhead. In
the spring and the fall it transits in between the
lowest winter point and the highest summer
point. My latitude is 44 degrees and 35 minutes
north. The optimal angle for any latitude is the
latitude angle itself. At my location, that's 44°
35' from the horizontal. I can leave this angle the
same for the whole year, but to take best advan-
tage of the sun's changing position, I adjust the
angle four times a year, halfway between each
solstice and equinox, which is about May 5,
August 5, November 5 and February 5. The rou-
tine is, on August 5 and February 5 the tilt is the
same as your latitude, 44° 35' for me . November
5, add fifteen degrees tilt to your latitude for the
winter solstice, for me 44° 35' plus 15° which
equals 59° 35'. May 5, subtract 15° from your
latitude for the summer solstice, for me, 44° 35'
Figuring the optimal
minus 15°, which equals 29° 35'. In practice I
angle for solar panels
round off everything and tilt my panels to 45° on
in the northern
February 5 and August 5, 60° on November 5,
hemisphere.
67
and 30° on May 5.
Solar Panels
The panels also have to be oriented facing true south. The fastest way to
calculate true south is go to the geomagnetic reference web site via the
Resources page of this book and use the magnetic declination calculator
to obtain what is called the magnetic declination for your longitude and lat-
itude. The magnetic declination is the difference in degrees, minutes and
seconds between magnetic north and true north. To use the magnetic dec-
lination figure to orient PV panels, use a compass to find magnetic north,
then add or subtract the magnetic declination according to your location.
Orienting the panel to true south and making adjustments for seasonal
variations will enhance hydrogen production.
Solar trackers
To increase production further, you can build or buy a solar tracker. A
solar tracker will follow the sun in its seasonal declination, as well as track
its daily east to west movement.
Trackers are a good addition to any solar hydrogen system. For a PV
system that has only battery storage, generally trackers do not improve
efficiency for much of the summer season. This is because battery banks
typically get fully charged in just a portion of a summer's day. But in a
hydrogen system, all the energy the PV panels can convert from sunlight
68
to electricity can be used.
Solar Panels
There are several types of trackers available including passive solar and
solar powered motor driven. They are not at all necessary for a system, but
they are an option to increase hydrogen production, if you want to experi-
ment with them. They are especially useful in very sunny climates. Their
value is probably not as great in cloudy climates, since the sunlight is more
diffuse. Some tracking systems take some outside energy to operate, and
thus must have their own PV power supply. Trackers also add to the com-
plexity of a system. Remember, anything mechanical requires more care
and maintenance, especially in harsh weather conditions and climates.
Mounting PV panels
PV panels can be mounted in many different ways. They can be secured to
racks that stand on poles, or be ground mounted, or put on roofs, for instance.
I would be cautious about any roof installation and do not advise it under most
circumstances. Panels need to be cleaned regularly, and they must be kept free
of snow and ice. If they are not easily accessible, this will not happen. I suggest
short pole or ground mounts, and if you live in snow country be sure the pan-
els are high enough off the ground to keep the panels out of the highest snow
cover  but low enough to remove the snow easily.
Panels should never be put in an area that is going to be shaded  not
69
even a little bit of shade. If a portion of a cell gets blocked from sunlight, it
Solar Panels
can take down most of the output of the entire panel and can cause other
more severe problems. Diodes used in a bypass fashion can alleviate
these problems, but they will add to system loss and should be avoided
whenever possible.
Make sure your PV panels are secure in high winds. Panels make great
sails and must be sturdily fastened to whatever mounting system you use.
Always leave plenty of air space at the back of the panels so that air can
circulate and cool them. They get very hot in use, and heat reduces their
efficiency and can stress the materials, especially your soldered connec-
tions, so do whatever you can to keep the panels as cool as possible.
If you can, avoid locating panels near dusty roads. Also, clean panel
surfaces at least once a week with plain water. Barely visible dust and
pollen can collect quite rapidly and affect the output of the panels.
Final design considerations
The final design of your panels should be based on your intended use.
Panels can be constructed in a number of different ways, and with variety
of materials.
For this particular project I did not build the ultimate panel, as I wanted
70
to use up materials I had on hand, such as the aluminum backing
Solar Panels
material. This use limited the panel output because it limited the number
of cells that could fit within the frame. If the materials on hand had not
been an issue, I would have made two panels with an output around 2.5 to
3 volts, or one panel with a 4 to 6 volt output. However, my yankee sensi-
bilities ruled and I figured the smaller size would be adequate for a proto-
type to run the electrolyzer.
The main consideration, if you build the electrolyzer in this book, is to try
to have 4 volts at about 20 amps for each electrolyzer. You can definitely
work with less, but due to circuit losses, varying sun intensity and so forth,
this is a good minimum to aim for.
Consider less than optimal conditions in your design. If you are looking
at long wire runs and diode use, plan this into the whole system.
If you wish to purchase ESPMs rather than build your own, be ready
for some sticker shock. While it is inexpensive to build these panels, it is
not inexpensive to buy them. The reason is that the photovoltaic industry
has not yet caught up with the idea of the coming hydrogen economy.
This is no surprise, since many PV companies are owned by large oil
companies, and for a variety of reasons they are not interested in pro-
moting small scale hydrogen production.
71
Solar Panels
The PV industry is geared for BSPMs, which means all of their tooling
and assembly line processes are set up to efficiently and economically
produce battery charging solar panels, but when the panel parameters are
changed, it becomes a special order and thus very expensive, even
though the materials used are identical. Hopefully this will change in the
near future. If you are involved in the photovoltaics industry, consider this
present void an economic opportunity, and a chance to open up the indus-
try to these possibilities. It would not take much research and develop-
ment to bring some good products for hydrogen production to market.
72
Electrolyzer
In a solar hydrogen system, the electrolyzer is the component which
changes electrical energy from the photovoltaic panels into hydrogen gas.
There are many types of electrolyzers: high temperature, high pressure,
low temperature and low pressure, and liquid electrolyte and solid elec-
trolyte forms. For solar hydrogen production, low to medium pressure, low
temperature liquid electrolyte electrolyzers are preferred. When compared
to the cost of high temperature, high pressure systems and/or solid elec-
trolyte systems, they are inexpensive to make, purchase and maintain.
Solid electrolyte PEM vs. alkaline
Solid electrolyte PEM (proton exchange membrane) electrolyzers can
be used in systems to avoid use of caustics as an added safety factor; and
where no one is available to frequently monitor a fluid electrolyte system.
PEM electrolyzers are much more expensive, and do not have the track
record that alkaline electrolyzers have in use. Although they are reported-
ly almost trouble free during use, they do pose problems in terms of cost
of replacement parts when they become inoperable. Failures in PEM elec-
trolyzers are usually membrane blow-outs or catalyst degeneration. Both
problems are costly to service with replacement parts.
Alkaline electrolyzers, on the other hand, use a very inexpensive elec-
73
trolyte and low cost electrodes such as nickel which are easy to obtain
Electrolyzers
and replace. Eventually, as lower cost alternatives to presently expensive
ionomers and catalysts become available, PEM electrolyzers will come to
the forefront and probably be more widely used than alkaline.
Cost was an important consideration for us in the development of our
experimental solar hydrogen system, so a PEM electrolyzer was discount-
ed in favor of a low pressure alkaline tank electrolyzer. High pressure and
high temperature electrolyzers were also evaluated. These were discount-
ed due to higher amounts of energy consumed and the need for more
intensive monitoring of the system.
Electrolyzer basics
An alkaline electrolyzer, as used in this system, is a simple electrochem-
ical device that disassociates water into its constituent molecules, oxygen
and hydrogen. This is accomplished by the application of very low voltage
and high amperage DC (direct current) electricity in an alkaline electrolyte
solution consisting of potassium hydroxide (KOH) and distilled water.
The term electrolyze means  to loosen with electricity. The term is
derived from the ancient Greek word elektron which means  amber. When
amber is rubbed, electricity is produced. The word lusis means  to loosen.
Thus, together they mean  to loosen by electricity.
74
Electrolyzers
Although an oversimplification, DC electricity flows in one direction only,
which is needed for the electrolytic process. AC (alternating current) elec-
tricity, such as used in our homes, flows in both directions at a specific
frequency which is not suitable for use in the process of electrolysis
unless it is rectified.
DC (direct current) electricity, which powers the electrolyzer, can come
from a variety of renewable sources, such as wind generators, photo-
voltaic panels and small hydro and geothermal systems. Batteries are a
common non-renewable source of DC current, as well as AC power sup-
plies with rectifiers that change the AC to DC.
An electrolyzer consists of two electrodes, usually made of pure nickel,
a nickel iron alloy, stainless steel, monel, or Raney nickel. One electrode
is connected to the positive source of the DC power supply and the other
electrode is connected to the negative source of the DC power supply.
The electrodes are immersed in a potassium hydroxide (KOH) solution
in a tank. The tank has collector tubes to carry off the generated gases.
Hydrogen is produced at the negative electrode and oxygen is produced
at the positive electrode.
75
Electrolyzers
There are many factors which determine the amount of hydrogen and oxy-
gen generated in the electrolyzer. The most important considerations are:
1. Percentage of KOH to water in the electrolyte solution.
2. Surface area of the electrodes.
3. Physical distance of the electrodes from each other.
4. Amount of DC current (amperage) applied
A higher ratio of KOH to water increases conductivity up to 29.4% of
KOH in solution. After this point, the resistance increases and there is no
point in adding more KOH.
The greater the surface area of the electrode the greater the gas pro-
duction will be. However, the greater the surface area, the greater the
amount of current that is needed to realize the full potential of the added
surface area. Hydrogen production, as it relates to current density, is cal-
culated by dividing the amount of current by the electrode area. This is
expressed as amps per area. Each electrolyzer has its own operating
parameters for its most efficient operation.
76
Electrolyzers
Porous alloy electrodes
Porous alloy electrodes of Raney nickel are often used in electrolyzers.
Raney nickel is produced by first making an alloy composed of 50% alu-
minum and 50% nickel. This composite is then treated with potassium
hydroxide, which eats away the aluminum and leaves a porous nickel
sponge material, known as Raney nickel after Murray Raney, the inventor
of the process. Electrolyzer electrodes of this material have a large sur-
face area due to their porosity and will produce more gas with smaller
electrodes, compared to electrodes made of materials with less surface
area, such as sheet or screen. Although Raney nickel is preferred, it
is more expensive than either sheet or screen. The porous texture that
creates a larger surface area also acts as a filter for small particles. The
sediment which forms reduces the active surface area over time, inhibiting
gas formation and thus efficiency.
Prepared surface flat plate electrodes
The surface area of flat plate electrodes can be augmented by sanding
or sandblasting the plate, thus improving gas production. Simple sanded
flat plates have less surface area than Raney electrodes would provide,
but they have more surface area than a simple flat plate. Besides being
77
Electrolyzers
economical, they can be reconstituted quite easily if needed, by resand-
ing, and so would last longer in operation. Use a very course grit sand
paper and apply the surface in a multidirectional pattern to get as much
extra surface area as possible.
If you want to experiment with these kinds of electrodes, a nickel-iron
alloy foil that is 80% nickel can be purchased from McMaster-Carr, such
as their #8912K24. This particular foil makes an excellent electrode and is
relatively inexpensive.
Mesh electrodes
Another alternative for electrodes is fine mesh screen. This provides
more surface area than the flat plate and is not as easily clogged up as is
the case with nickel sponge. Either 316 stainless steel, monel or nickel
screen can be used. Nickel screen is expensive, so the 316 or monel is a
more cost effective choice. The best choice is the monel, an alloy of 65%
nickel, 33% copper and 2% iron. This alloy has excellent corrosion resist-
ance in alkaline solutions and will last a long time.
78
Electrolyzers
Electrode spacing
The closer the electrodes are to each other, the more copious the gas
production. However, the closer the electrodes, the more risk there is of
mixing gases, even if one uses membrane separators.
Some commercial electrolyzers sacrifice gas purity for electrical effi-
ciency with closely spaced electrodes, only to have to add expensive
purification equipment at the end of the process, which negates any cost
efficiency at the production point. The more cost effective solution for
purer gas production is to err on the side of separating the electrodes a
bit more, and sacrifice a certain amount of electrical efficiency.
The more current, the more gas will be produced, within reasonable lim-
its. All electrical systems have voltage and current parameters that, if
exceeded, will deteriorate production or destroy the equipment. The cur-
rent carrying capacity of the electrodes is a factor here.
A minimum input of about 1.29 to 1.49 volts is needed to initiate the
process of separating the hydrogen and oxygen that constitute the water
molecule. Voltage applied to the electrodes is usually in the range of 2.5
volts to 6 volts. The electrolyzer in operation will usually draw between 1.7
to 4 volts.
79
Electrolyzers
The electrolyte
Potassium hydroxide (also known as caustic potash) is a strong elec-
trolyte. This means that it is essentially 100% ionized in solution and thus
is a good conductor of electricity. When the positive and negative poles of
the electrolyzer are connected to the power source, hydrogen ions com-
bine with electrons at the negative electrode to form hydrogen, and
hydroxy ions give up electrons at the positive electrode, releasing oxygen.
Twice as much hydrogen is generated as oxygen, since the water mole-
cule contains two hydrogen atoms for every oxygen atom.
Other electrolytes, such as sodium hydroxide, can be used, but they are
not as conductive as KOH. Acids such as sulphuric acid can used as elec-
trolytes, but this is more corrosive to the electrodes, and the added wear
and tear on various components does not justify its use.
Essentially KOH is the best choice for alkaline electrolyzers.
KOH can be purchased, or made if you have a source of hardwood
ashes. The making of lye used to be a common chore in most households.
If you are interested in making your own KOH my book Build Your Own
Fuel Cells contains complete illustrated instructions. If you buy KOH from
a chemical supply house you will find the following table helpful.
80
Electrolyzers
Potassium hydroxide (KOH) solution strength
Specific Percent Lbs. per Specific Percent Lbs. per
Gravity KOH US Gallon Gravity KOH US Gallon
1.0083 1 0.0841 1.1493 16 1.535
1.0175 2 0.1698 1.159 17 1.644
1.0267 3 0.257 1.1688 18 1.756
1.0359 4 0.3458 1.1786 19 1.869
1.0452 5 0.4361 1.1884 20 1.983
1.0544 6 0.528 1.1984 21 2.1
1.0637 7 0.6214 1.2083 22 2.218
1.073 8 0.7164 1.2184 23 2.339
1.0824 9 0.813 1.2285 24 2.461
1.0918 10 0.9111 1.2387 25 2.584
1.1013 11 1.011 1.2489 26 2.71
1.1108 12 1.112 1.2592 27 2.837
1.1203 13 1.215 1.2695 28 2.966
1.1299 14 1.32 1.28 29 3.098
1.1396 15 1.427 1.2905 30 3.231
If you make your own KOH, use this handy table to determine the
specific gravity of the solution when you take a hydrometer reading. This
will indicate whether to boil the solution down more to strengthen it, or add
more distilled water to weaken it.
81
Electrolyzers
I make my own KOH and do not bother to boil it down. I simply refill my
reservoir with more KOH solution rather than distilled water. The KOH,
when I first make it, comes out at about a 12% solution. In the electroly-
sis process, the solution becomes stronger as the water disassociates and
is used up, leaving the KOH behind. Once the solution in the electrolyzer
reservoir is at the right specific gravity, all that is necessary is to add dis-
tilled water as the water in the electrolyte disassociates and is used up.
If you start with the exact specific gravity that you want when you first
fill the reservoir, then afterward, simply replenish the reservoir with dis-
tilled water to the same level of your first fill. KOH is lost over time, but it
is mostly the water that is used up in the process. Also, be sure to use dis-
tilled water only. Well, tap, and spring water contain far too many
unknowns (minerals, organic particles, etc.) that will cause problems with
the electrolyzing process and gum up the electrodes.
Never mix dry purchased KOH into the water in the reservoir  the
process generates too much heat. Always mix KOH into the water by put-
ting a little bit in at a time, very slowly. Do not mix by pouring water onto
the KOH.
82
Electrolyzers
Use only plastic or stainless steel buckets or containers. Do not use alu-
minum pots or utensils for mixing or holding KOH solution!!! And, always
let the solution cool down before refilling the reservoir.
For best conductivity, the solution should be about 29.4%, which means
it would have a specific gravity of about 1.28. This would require 3.231
pounds of KOH for one gallon of distilled water.
This particular specific gravity is not necessary, and a milder, less con-
ductive solution of about 12% will work. For optimal performance howev-
er, you will want to work with a 29.4% solution. To check the specific grav-
ity of a solution, use a hydrometer such as those designed for testing the
specific gravity of battery electrolyte. Hydrometers can be purchased at
any auto parts or hardware store.
Safety
If you work with KOH, never forget that it is extremely caustic and can
cause severe burns and blindness if not handled properly. Please do not
ignore these warnings. Make sure you have eye protection such as safety
glasses or a safety face shield. Either can be purchased inexpensively at
most hardware stores. Always have all your skin completely covered with
83
Electrolyzers
protective clothing and rubber gloves when working with and around KOH;
and study and follow all MSDS recommendations. (See Resource page for
URLs). Always diligently follow all safety precautions when handling KOH,
and keep it out of reach of children and pets.
In this particular system, gas pressure is supplied by a hydrostatic column.
This is simply a filler tube with a reservoir on top at a height of five to eight
feet, depending on how much pressure you want in the system. Because the
reservoir is at eye level or above, it must be very well secured. You absolute-
ly do not want nasty KOH raining down on you during your experiments. Hose
clamps should be used to secure the tubing to the reservoir so that hoses
cannot slip off and squirt KOH all over the place.
Water for the electrolyzer
The electrolyzer that you use in your system needs distilled water.
Distilled water is for the most part contaminant free. Municipal tap water
usually contains chemical additives, and well water contains dissolved
minerals which, over time will encrust the electrodes and impede the
action of the electrolyzer, so neither of these is suitable for electrolyzers.
Rain water or melted snow are second choices to distilled water, although
there are still some contaminants in these as well. The water for an elec-
trolyzer should be distilled, whether you buy it, or make it yourself.
84
Electrolyzers
Solar water distillation
One of the best and most eco-
nomical ways to distill water is
with a solar water distiller. They
are easy to make and, if you
have the space, can supply some
or all of the distilled water need-
ed for the electrolyzer system.
The illustration at right shows a
Simple solar water distiller
very basic solar water distiller.
This is a very simple device that
evaporates the water, leaving any mineral constituents behind. The water
condenses on the glass and then drops into the trough and then out
through the tube into a collecting bottle.
The glass cover can be a used patio door or any other piece of glass
(about door size). The box is easy to construct, and can be made of wood
or metal. The water trough inside the distiller can also be made of wood,
metal, or heat resistant plastic. Tubing should be silicone, and reservoirs
and collection bottles etc. can be plastic or glass. The inside of the box
85
Electrolyzers
has to be lined with a UV (ultraviolet) resistant and non-contaminating
material, and needs to be black in color so that it absorbs heat and is leak
proof. Black silicone caulk can be used to cover the entire inner surface,
or some type of rubber or plastic liner that will not degrade can be used.
The glass top should lie flush on the box with a tight fit to keep the heat
in so that water vapor forms on the glass.
Volunteers In Technical Assistance (VITA) has a booklet by W. R. Breslin
titled Solar Still (available through PACT Publishing) which has instruc-
tions for building a solar heated distillation system.
The P41 electrolyzer
The P41 is a simple electrolyzer specifically designed to provide maxi-
mum output with the smallest footprint possible with off the shelf compo-
nents. As with every part of this project, the important factors were ready
availability of components at low cost, and ease of working with the com-
ponents, so that anyone with average skills could build a device that would
be comparable with commercial equipment.
For this electrolyzer, the primary goal was to produce enough hydrogen
to run a real time stationary fuel cell system within the smallest possible
space without the gases intermingling.
86
Electrolyzers
Design for renewable energy power sources
An important design consideration was to maximize the efficient use of
energy input from renewable energy power sources. Most commercial
electrolyzers are designed to run off utility power grids with a rectified DC
source. The electrode materials used in these electrolyzers reflect that
particular type and quality of power source.
Making intermittent power efficient
DC power from renewable sources such as photovoltaic panels is deliv-
ered intermittently. Depending upon where you are, there are very sunny
clear days, but there are usually more days when the sun is behind the
clouds for five minutes then bursts out in full sun only to be behind anoth-
er cloud in another five minutes. This varying amount of sunlight will cause
the disassociation of hydrogen and oxygen from water to proceed at a
faster or slower pace according to the unsteady flow of DC power from the
PV panels. This causes sharp spikes and troughs in gas production. This
is not a problem per se, but it led me to think of nano-nuances in regard
to renewable energy delivery systems.
A minimum voltage of about 1.23 V at about 77°F will disassociate water
into hydrogen and oxygen, but this process absorbs heat. At a voltage of
87
about 1.49 at around 77°F, no heat is absorbed. Higher voltages than 1.49
Electrolyzers
release heat during the disassociation process. Elevated operating tem-
peratures increase the efficiency of the electrolytic process because less
electrical input is needed.
Electrode materials
Experiments with a variety of materials led to the conclusion that monel
is a perfect choice for electrodes in an electrolyzer designed for intermittent
power supplies. It has an electrical resistance of about 42 micro ohms-cen-
timeters at 20°C. This is much higher than, for instance, nickel, which has a
resistance of 11. With higher resistance, more heat is generated as the cur-
rent passes through the electrodes and electrolyte solution.
Thermal flywheel
For conventional electrolyzer design, monel would not be considered to
be as efficient as other materials  the heat generated would be consid-
ered wasted energy. However, for intermittent renewable energy systems,
monel is a good choice because thermal input dissipates at a much slow-
er rate than electrical input. Thus, the thermal energy is still feeding the
process when, for instance, the sun is momentarily obscured by a cloud.
A  thermal flywheel is created that retains the energy gained from the
photons as heat, which lowers the threshold for the electrolytic reaction in
88
terms of how much voltage and current is needed. When the sun is behind
Electrolyzers
a cloud, less electricity is generated, but the heat retained in the elec-
trolyzer requires less electricity to continue the reaction at a more active
level than without the additional heat.
Supercapacitors
Another approach to consistent power delivery from renewable sources
is to consider the use of supercapacitors placed between the renewable
power source and the electrolyzer. These may or may not be viable, and
we have not yet researched this possibility. Supercapacitors are basically
a cross between capacitor and battery technology. They use electrodes,
and a liquid or organic electrolyte, but they store energy by static charge
rather than by electrochemical means. They can be cycled millions of
times, and have a recharge time of seconds. Supercapacitors might also
be viable to enhance peak load performance on the fuel cell end.
Micro and macro electrode surface considerations
The next design consideration was surface area. The greater the surface
area of an electrode, the greater the gas production will be for a given
space.
89
Electrolyzers
Raney metal surfaces
At the present time, Raney metal structures expose the greatest amount
of surface per area. (Nano-tube structures will soon surpass Raney effi-
ciencies, but right now these nano-tube electrodes are very experimental.)
However, as stated earlier, Raney structures have the nasty habit of act-
ing as extremely fine porous filters that gum up. This would reduce the
efficiency of an electrolyzer incrementally over time. Also, the micro
porous Raney surfaces seem to retain gas for longer periods of time in
pockets, which blocks the reaction and renders a portion of the electrode
surface useless. This creates a higher active threshold, which is accept-
able for industrial systems that run off consistently supplied grid power,
but is not particularly suitable for intermittent renewable power supplies.
Another disadvantage is that Raney surfaces are more expensive than
other alternatives. They are mainly used in industries where cost of elec-
trode replacement is not a concern. They make excellent electrode sur-
faces, but keep in mind the degeneration that can occur in a not too per-
fect environment, and that replacement of electrodes should be consid-
ered in your design.
Raney electrodes did not have the profile we were looking for. We need-
ed a very inexpensive material that was rugged and would last almost
90
Electrolyzers
forever. 200 monel mesh seemed to be the right answer. It is extremely
resistant to corrosive environments, provides a larger surface area than flat
or surface sanded plates, and it does not retain gas bubbles for long peri-
ods of time so that the same surface could react more frequently
Electrode shape
After considering the micro surface, the next task was to design the
macro surface, which is the shape of the electrodes themselves. After much
consideration, we came up with a star-pleated design for the negative elec-
trode inside a plain cylinder positive electrode. In this way, we could match
the surface areas of the two electrodes in a small amount of space. There
are detailed illustrations of the electrode design later in this chapter.
We wanted to have as much macro surface as possible, but at the same
time did not want to overdo it. We expected better than average gas pro-
duction and had to consider the volume of gas generated in relation to the
size of the electrolyzer tank.
91
Electrolyzers
Hydrostatic pressure
The gas pressure for the electrolyzer system is provided by a hydro-
static column. This can be simply a vertical tube, the height of which pro-
vides pressure to the system as described in this formula:
1 psi = 2.31 ft. water column,
or, to put it another way:
.036127 psi per inch column of water.
Thus, a tube 5 high would provide a pressure of 2.17 psi, a tube 6 high
would provide a pressure of 2.60 psi, a tube 7 high would provide a pres-
sure of 3.03 psi, a tube 8 high would provide a pressure of 3.47 psi, and
so on.
The hydrostatic column is a simple way to regulate the pressure up to a
certain point. The Romans used this principle to apply pressure to the
water supplied through pipes from their aqueducts This method is suitable
and cost effective for stationary systems that require only a small amount
of pressure to operate. It is also somewhat safer than pressurized systems
when a caustic electrolyte is used.
92
Electrolyzers
An enclosed pressurized system would be the next step for experi-
menters who wish to design a more compact system. More compactness
can be achieved by designing the electrolyzer system to withstand higher
operating pressures, as you require. Such a system would need the addi-
tion of relief valves to guard against over pressure, and a pressure gauge
and float valve to feed the electrolyzer with KOH solution or distilled water.
The hoses in the system also would have to have sufficient rating for the
pressure, and would require hose band clamps, and so on.
To keep things simple and inexpensive, we used the hydrostatic column
to provide gas pressure to the system. This low pressure approach per-
mits more flexibility in that components can be quickly added or subtract-
ed to refine or change the system design as a whole. The components are
also less expensive, which is useful for building prototypes and learning
the basic principles of operation. Once you get used to their operating
quirks and understand how these systems operate by observation, then
you can upgrade them, and graduate to more expensive components 
and make pressure design changes with some experience behind you.
This more studied and slower approach will give you the skills to go on
and design some very compact and sophisticated systems.
93
Electrolyzers
Building the P41 Electrolyzer
Tool list
All should be available from local hardware or electronics stores.
Long nose pliers
Epoxy glue
Scissors
Long screwdriver, 6" or more
Metal cutters
Soldering iron, 15 watt, pointed tip
Caulking gun
Screwdriver
1 3
Drill bits, /4 " and /8 "
Hacksaw, and other types of saws to cut metal and other items
Drill, either handheld or drill press
94
Electrolyzers
Materials list
All should be available from a local hardware store,
unless otherwise noted.
11 Stainless steel nuts for 10/24 screws.
1 PVC pipe, thin wall 3" ID, 91/8 " length.
1 3
Nickel alloy foil tabs, 010 thick. 5 pieces each /2 " to /4 " long, and
nickel alloy washers 1" diameter McMaster-Carr, part #8912K241,
comes in a sheet 4"x 12".
1
Polypropylene white felt, /16" thick, 73/4 " long x 515/16" width.
McMaster-Carr, part #88125K11. Rolls are 12 x 72".
1 PVC pipe coupler 115/16 ID, 27/16" long.
1 PVC pipe cap 1" ID.
1
6 Stainless steel washers, /2 " diameter to accommodate 10/24 screws.
1 1
4 Silicone rubber spacers /8  thick x /4 " wide X 6" long. 2 silicone
rubber washers 1" diameter. McMaster-Carr, part #7665K22,
comes in a 1 wide strip, 36  long.
1 Rubber o-ring 1"ID.
95
Electrolyzers
Materials list (continued)
Screws  pan head stainless steel, two 10/24x 1", and one 10/24x
11/4 ".
Monel standard grade woven wire cloth, 200x 200 mesh, .0021" wire
diameter, 12"x 12" sheet. McMaster-Carr, part #9225T264.
2 PVC pipe caps for 3" thin wall PVC pipe.
3
Barbed hose connectors, two /8  and one 11/4  .
Clear silicone rubber caulking.
96
Electrolyzers
Building the P41
The P41 consists of very
few parts It is inexpensive to
build, is suitably matched for
renewable energy inputs
from photovoltaic panels or
wind generators, and it has
a high volume of gas output
for its size.
Electrolyzer tank
The electrolyzer tank is
easily constructed from thin
Parts for the P41 electrolyzer,
wall PVC pipe and pipe caps.
felt separator not shown.
Cut one piece of 3" thin wall
PVC pipe to 91/ 8" length.
This will be the body of the electrolyzer chamber.
Positive electrode assembly ports
Drill a hole to accept a 10/24 screw in the pipe 3" from one end of the
97
pipe. Drill a similar size hole directly opposite that hole on the other side
Electrolyzers
of the pipe. You will then have two holes
drilled to accept 10/24 screws on opposite
sides of the pipe that are 3" from one of
the pipe ends. These holes are the entry
ports for the positive electrode assembly.
Gas exit port cap preparation
Take one of the 3" pipe caps and drill a
3
/ 8" hole in the center of the cap. The hole
is the hydrogen gas exit port, and the cap
Prepare the gas port cap
will be the top of the electrolyzer. Center a
115/ 16" pipe coupler inside the cap (see
bottom photo at right), and use a fine tip
marking pen to outline the coupler s outer
circumference inside the PVC cap. This cir-
cle is a gluing guide for attaching the cou-
pler to the cap.
1
In the same cap, drill another / 4" hole
between the inner wall of the pipe cap and
the circle outline you drew in pencil. This
98
hole is the oxygen gas exit port.
Electrolyzers
Take a 115/ 16" ID (inner diameter)x
27/ 16" long PVC pipe coupler, and coat
one edge with epoxy. Coat the circle out-
line on the inside center of the cap with
epoxy also. Align the coupler with the out-
line of the circle and press it. Put extra
epoxy around the outside and inside edges
of the 115/ 16 gas takeoff to ensure a very
good seal. A tight seal is critical so that the
hydrogen does not leak to the oxygen col-
Side view of the
lection side, and visa versa.
gas port cap
Let the cap and coupler dry for 24 hours.
Then, apply epoxy around the barb and
3
around the entry port hole, and place the / 8" cut off barb in the center.
The barb should not protrude into the cap. It should be flush with the
inside wall. Leave the barb long enough on the top, outside of the cap, to
have as much as possible to fit the hose over. If it is too short, the tube
will not have a good gripping area when it is placed on the barb. Repeat
the same process for the oxygen take off. Let it dry for 24 hours.
99
Electrolyzers
Be careful drilling holes for exit
barbs in these thin walled pipe
caps. If you drill them a tad too
large or at a slight angle, the barb
will be floppy when seated on the
cap. The fit should be as tight as
possible.
Separator preparation
Cut a piece of polypropylene felt
Use a soldering iron to bond
73/ 4" X 515/16". Set up a 15 watt
the seam of the rolled up felt
soldering iron with a new tip. I used
a pointed tip, but you can
also experiment with what
are called chisel or screw-
driver type tips, or try a
higher wattage iron and
see how that works. A heat
gun or wood burning set
element might work quite
well also.
100
Electrolyzers
Use the soldering iron as a heat
welder to bind the edges of the felt
together to form a tube. The tube will
be 73/ 4" long. The circumference of
the tube will be formed from the
515/ 16" width of the felt. To form the
circumference, fold two edges to the
center of the width and make sure
the edges abut well and evenly (see
illustrations previous page). Hold the
Separator tube after bonding
edges together on one end and apply
heat to bond, by stroking the soldering iron across the seam where the
two edges abut. This will melt the plastic and bind them together.
Move down the piece and stroke across the edges as you go, welding
the entire length of the tube in this manner. Stroke the iron quite quickly
and lightly over the seam. If you hold the iron a second too long, you will
melt too much plastic and ruin the piece. Be sure to make each welding
stroke close to the previous one so that there are no gaps in the weld. If
you hold the piece of felt up to the light you will be able to see any gaps.
If there are gaps, go over the gap area lightly until it is closed. It is impor-
101
tant to make sure that no gaps exist so that the hydrogen and oxygen
Electrolyzers
gases do not mix. Practice the technique
on some scrap pieces first, and then pro-
ceed to the final piece. This felt tube fits
inside the 115/ 16 pipe coupler on the
hydrogen gas exit port.
Positive electrode assembly
preparation
Cut a 6"x 91/ 2" inch piece from the Monel
Glue rubber spacers
wire cloth according to the layout on page on the inside of
the electrolyzer tube
108. This will be the outer oxygen electrode
and will follow closely the inside surface of
the PVC pipe.
1
Cut four silicone rubber rib spacers / 8"
1
thick, by / 4" wide, by 6" long. Cement
these strips with silicone or epoxy to the
inner walls of the electrolyzer chamber
more or less equidistant from each other,
with one end of the strip lining up with the
bottom edge of the pipe. Be sure not to
102
cover over the positive electrode port
Electrolyzers
holes. These strips keep the wire mesh away from the wall of the pipe so
that gas can be created on the outward side of the positive electrode.
Washers
The next step is to make washers for the positive electrode assembly.
These are two nickel alloy washers and two silicone rubber washers. Use
a 1" diameter washer as a model to trace the circumference and center
hole onto the rubber and the nickel alloy sheet. Use a punch to cut out the
center hole in the rubber sheet, and a punch or drill to cut out the center
hole on the nickel alloy. The circumference of the rubber sheet can be cut
with an exacto knife and the circumference of the nickel washer can be cut
with tin snips.
After the rib spacers glue has set, roll the 6"x 91/ 2 monel mesh screen
into a cylinder (6 long) that can be slipped into the electrolyzer body.
When the piece of mesh is fully inside the tube
on the ribs, adjust it to even it out. There
should be an overlap. This is the tabbing junc-
tion. Adjust the monel screen so that it is even,
Secure the rolled screen at the correct
103
diameter with crimped nickel tabs
Electrolyzers
round and sitting on the ribs well,
but not touching the walls of the
pipe. This is the proper circumfer-
ence to allow gas flow between the
mesh and the walls of the pipe.
Cut crimp tabs from the nickel
alloy. These crimp tabs secure the
Slip the cylinder back into the pipe
overlapping edges of the Monel
mesh on both ends of the cylinder
3 1
(see illustration at right). Cut the crimp tabs about /4 long and about /8"
wide with tin snips. They can be wider and longer if desired. Making sure that
the mesh cylinder retains its form, slip the cylinder a little bit out of the pipe
and use pliers to crimp the tab on the overlapping mesh to secure the two
edges together. Pull the mesh totally out of the tube and secure the other end
of the cylinder with a crimp tab.
Slip the cylinder back into the pipe and position the overlapping seam
over one of the electrode port holes. Inside the tube, place your fingers over
that hole, and press the mesh against the pipe.
104
Electrolyzers
Insert a pointed object such as a small
drill bit through the hole from the outside
and push it through the mesh. The drill bit
will move the mesh strands aside to create
the hole rather than breaking them. This
hole needs to accommodate a 10/24 screw
that will be inserted through it, so be sure it
is the correct size. Remove the drill bit and
replace it momentarily with the 10/24 screw
to hold the mesh in position while you make
the hole on the opposite side of the cylin-
der. Create this second hole the same way
Insert a drill bit through the
as the first. Take care doing this so that the hole in the pipe wall to
make a hole in the positive
cylinder retains its shape and the holes are
electrode screen
correctly aligned.
Slip out the screw and the mesh cylinder.
Take a nickel alloy washer, place it on a 10/24 screw and insert the
screw through one of the holes in the mesh cylinder, with the head of the
screw and the nickel washer on the inside of the cylinder. Then slip a sili-
cone rubber washer on the outside of the cylinder over the screw.
105
Electrolyzers
Do the same with the other port
hole. Back off the screws a bit to fit
the cylinder back into the pipe, then
push the screws through the port-
holes so that they extend to the
outside of the PVC pipe.
The next step is to screw on a
nut. Apply a good quantity of sili-
Insert the electrode connector screws
cone to seal the hole and the nut to
from the inside, though a nickel
the hole and pipe before you tight-
washer, then the Monel screen,
en the nut to the pipe wall. Tighten
then a rubber washer before they go
through the pipe wall.
the nut and apply more silicone to
make sure the nut is sealed well
against the hole. Repeat for the
other hole.
During this process, try to be
neat and not get silicone on the
screw surfaces where the electrical
connections will be made, as the
silicone will insulate the electrode
106
Electrolyzers
from the power source and the current will
not flow. If you mistakenly get some on the
screw body where the contacts will be,
clean it off thoroughly to be sure the sur-
faces are electrically conductive and not
impeded by silicone.
When the silicone has dried thoroughly,
slip another nut on, and tighten it against
the first nut, then slip the two washers on
The installed positive
and the other nut. The connecter will be
electrode connectors
attached to the electrode assembly
between the two washers.
Finally, test for continuity with a multimeter. If you are not getting conti-
nuity, check to see if there is silicone between the pan head of the screw
and the washer and screen. As mentioned above, silicone will insulate the
connection and the current will not flow. If you apply glue as neatly as
possible in the first place, there shouldn t be any problems.
107
Electrolyzers
Electrode cutting layout
Layout for cutting the positive and negative
108
electrodes from a 12"x 12" piece of monel screen
Electrolyzers
Negative electrode
assembly preparation
Cut the monel wire mesh
for the negative electrode to
dimensions as shown in the
diagram on the previous
page. This will give you a 6"x
3
93/ 8" piece with a / 4" wide
Mark the fold lines for the
pleats on the screen
tab at the bottom. This piece of
mesh will be folded into pleats.
To make the pleats, mark the
fold points on each side of the
screen (see illustrations). The
3
fold points should be / 8" apart.
Begin folding using a ruler along
109
each fold line as shown in the illus-
Use a metal ruler to make sharp
creases for each fold.
110
Electrolyzers
trations on the previous page. When all the folds/pleats have been made, fold
the tab back a long the edge of the main piece towards the pleats and crease
the fold, as shown below (step 1), then finish the tab by folding as shown in
111
Electrolyzers
steps 2-4, above. Roll the ends of the
pleated mesh together with the tab com-
ing out from the inside of the roll, overlap
the edges as shown at left, then crimp
the overlapped end pleats together with
nickel alloy tabs. With a drill bit or other
small pointed object, make a hole in the
center of the tab for a 10/24 screw (step
4, above).
112
Electrolyzers
Preparing the bottom cap
Drill a hole in the center of the pipe cap
for the bottom cap to accept a 10/24
screw. Then, center the 1" PVC cap inside
the cap, and draw an outline with pencil.
The 10/24 screw hole should be dead
The pleated and rolled
3
center within this outline. Drill a / 8" hole
negative electrode, ready to
between the inner wall of the 3" PVC cap install in the bottom cap.
and the circle outlined for the 1" PVC cap
3
(see illustration at right). This / 8" hole is
the electrolyte entry port. Cut and epoxy
the 3/ 8" barb connector into the hole.
Drill a hole in the center of the 1" inner
diameter (ID) PVC cap to accommodate a
10/24 screw. Make a 1" nickel alloy wash-
er. Apply a small amount of silicone
Inside the bottom cap,
showing guide outline, center screw
113
hole, and electrolyte entry port
Electrolyzers
around the area outlined for the 1"
ID pipe cap on the inside of the 3"
pipe cap. Keep the silicone away
from the center hole which will
accommodate the 10/24 electrode
connector screw.
Apply silicone to the outside top
of the 1" ID pipe cap. Keep the sili-
cone away from the electrode hole
so that when the screw connector
goes through this hole, it does not
get silicone on its threads, which
would compromise the electrical
connection. Now, push the 10/24
screw electrode connector through
the wire mesh (see illustration)
Assembling the connection
Place the 1" ID cap in position in
for the negative electrode
the center of 3" electrolyte entry
cap and seat the silicone covered
114
Electrolyzers
surfaces. Drop the 1" nickel alloy washer into
the 1" ID cap. It should fit in snugly.
If the entry holes for the 1" pipe cap and 3"
pipe cap are tight for the screw, start the
screw in and then back it out. This will cut
threads into the pipe. If the holes are not that
tight the screw will push through without
threading. Ideally, the screw should go in
tightly.
Take a 6" long screwdriver and insert it
down through the center of the wire mesh
electrode and push or screw the 10/24 elec-
trode connector through the holes with the tab
mesh connected, as shown. This will seat the
mesh tab onto the nickel alloy washer, and
secure the mesh electrode to the cap. Apply
silicone around the screw and screw hole
where it exits the cap. Use enough silicone to
form a good seal for the nut. Put on a nut and
Use a long shafted screw
tighten it. Apply more silicone around the nut
driver for the electrode
115
connection screw.
where it touches the PVC, again being careful
Electrolyzers
not to get silicone on the part of the screw above the nut, as it will inter-
fere with a good conductive connection. Inside the cap, apply more sili-
cone where the 1" cap comes together with the 3" cap. You can smooth
this seal with your fingers.
The next step is to put some silicone around the rim of the 1" cap that
seats the negative mesh electrode. Take a 1" rubber o-ring, slip it over the
mesh and slide it down to the 1" cap, seating it on the cap. The rubber o-
ring will steady and keep the mesh electrode straight. It is important for
the mesh electrode to stand straight up and not be seated at an angle.
Let the silicone dry. Test the mesh electrode for looseness in the 1" cap.
If it stands straight up it is ok. If not, apply a drop of epoxy where a pleat
or two touches the 1" PVC pipe and let it dry. While drying, make sure the
mesh screen stays straight and vertical.
Installing electrolyte entry head/negative electrode assembly
Coat the outside of the pipe and inside of the top pipe cap with silicone,
as done with the entry head/bottom cap. Be sure that the negative elec-
trode is secure and standing straight upward. Take the felt separator tube
and slip it into the 115/ 16" hydrogen exit port attached to the exit head.
Then, slip the other end of mounted felt separator tube down over the
116
Electrolyzers
negative electrode, and seat the cap, finally
enclosing the electrolyzer chamber. Coat
the seam between the cap and the elec-
trolyzer chamber tube with silicone where
they meet and smooth out with your finger.
Installing gas port exit cylinder
head and gas separator
Coat the outside top edge of the pipe
with silicone around the circumference for
Installed positive and
the length that the pipe cap will travel when
negative electrodes.
seated. This will be about 11/ 2". Coat it lib-
erally. To form a gas seal, the area must be
thoroughly covered. Then, coat the rim of the pipe with silicone, and then
coat the pipe cap walls with silicone around the inner circumference. Make
sure the surfaces are well covered.
Do not use so much silicone that it will be messy and squeeze out all
over the place, but at the same time be sure that the coverage is thorough
to make a good gas seal.
117
Electrolyzers
Seat the pipe cap, sliding the felt tube down over the
negative electrode. Make sure the cap is pressed all the
way down. Use a flashlight to see if any silicone is imped-
ing the oxygen port. If it is, remove it with a long stick. I
use wooden BBQ sticks from the grocery store. They are
handy for gluing in long tight spaces.
Congratulations, you have just constructed a very good
quality, high performance electrolyzer!
Hydrostatic testing
To test the integrity of the silicone bonding of the electrolyzer's compo-
nents, conduct a few leak tests. Use a Rubbermaid® container or a bucket
filled with water. Connect pieces of hose to each of the ports (hydrogen out-
let, oxygen outlet and electrolyte inlet). The hose pieces should be long
enough so that you can immerse the electrolyzer completely in the water
and have the open tube ends above the water with no liquid flowing into the
electrolyzer through any of the three ports. Block off the ends of two of the
tubes, blow through the third open ended tube and note if any bubbles come
from the electrolyzer. If they do, note where the leak is. Take the electrolyz-
er out of the water and dry it, then seal any leaky spots with more
118
Electrolyzers
silicone. When the silicone dries, leak test it
again, note any further leaks and apply sili-
cone as before until all leaks are corrected.
Power supply connections
The power cord coming from the power
supply should be a 12 gauge, or preferably a
10 gauge zip or other wire if you are using
ESPMs. Other wire can be used, as long as it
is rated for this use.
The connectors crimped on the end of the
wires that connect to the electrolyzer should
be ring type connectors so that they will not
slide off if the nuts loosen. To connect the
ring connector to the electrolyzer, slip one
washer against the nut on the electrolyzer
tube, slip on the ring connector with wire
attached, put another washer on and then
tighten a nut over that.
The completed P41 electrolyzer
119
Electrolyzers
This particular electrolyzer has two positive
pole connectors on either side of the body of the
electrolyzer, and one negative connection on the
bottom. When the wires are connected securely,
coat the connectors with silicone to insulate them.
You can also use a rubber dip compound or liquid
electrical tape for this.
Testing the electrolyzer with KOH
When the electrolyzer has satisfactorily passed
the leak tests, connect tubing to the KOH port at
the bottom of the electrolyzer. This tubing should
be long enough to come up above the top of the
electrolyzer and should be secured to the elec-
trolyzer with a rubber band or cord to hold it in
Test set-up
place. (See illustration) Push a funnel inside the
opening of the tube and use this as a fill funnel for
your KOH. Attach tubes to the two gas exit ports and support them above
the electrolyzer. Fill the electrolyzer with KOH solution through the funnel.
Wear protective clothing, gloves, and you absolutely must wear safety
glasses or a safety face shield. Fill the electrolyzer just a little above the
120
Electrolyzers
top so that you can see the KOH solution rising
in the tube just above the top of the electrolyzer
in all three tubes.
The tubes from the gas ports should be long
enough to put them in a glass of water. Connect
the electrolyzer to the power source and wait for
the bubbling action to start in the glass. If every-
thing appears to be working, empty the elec-
trolyzer carefully and the testing is complete.
You can run distilled water through the elec-
trolyzer after testing and emptying out the KOH,
to flush out any residual KOH solution.
Setting up the electrolyzer
Level the electrolyzer for operation. Attach the
hydrostatic column/electrolyte feed tube to the
feed port at the bottom of the electrolyzer. The
length of the feed tube/column will depend on
Hydrostatic column
how much pressure you want. Any length from
set-up for operation
3.5' to 8' will suffice. You can add a set of gas
121
Electrolyzers
check loops to connect the hydrogen, oxygen and
hydrostatic column tubes together (see illustration
at right). This is not necessary, but if the elec-
trolyte runs low in the tubing, this will give the
gases an outlet through the hydrostatic column.
Set up the hydrostatic column along a wall or
other support structure, and attach the tubes with
pipe hangers. Attach the top of the column tube to
a reservoir.
Electrolyte reservoir
The reservoir will be filled on a regular basis to
maintain hydrostatic pressure and provide elec-
trolyte and/or distilled water to the system. The
reservoir should have a cover or lid to keep out
debris and to slow evaporation. Make a small hole
in the cover so that a vacuum is not created in the
reservoir.
Hydrostatic column
set up with
The reservoir can be anything from a 32 oz.
gas check loops
recycled yogurt container to a large custom built
122
Electrolyzers
tank. Rubbermaid® containers, and small tanks
from Aquatic Eco Systems made from
polypropylene or polyethylene are good candi-
dates. You can also purchase one piece con-
tainers with sealed partitions so that you can
have several reservoirs in one container. Round
containers can be hung on a wall/support struc-
ture with metal flower pot holders or large size
pipe hangers. Make sure to use materials that
are compatible with KOH. Do not use metal con-
tainers unless they are stainless steel.
Whatever type of container you use, drill a
hole in the bottom, and insert and epoxy a barb
or threaded/barb connector through the hole.
This connector attaches to the column/feed
Electrolyte reservoir
tube. You may also want to glue a small piece of
mesh over the hole in the reservoir to act as a
filter. Let the glued connector dry.
Put a hose clamp over the tube before connecting and then slide the
hose over the barb and tighten the hose connector. It is important that this
123
Electrolyzers
hose connection be very secure so that the tube with KOH will not slip off,
and rain caustic KOH down upon you.
Set up the exit port tubes
The exit port tubes need to rise at least 1' above the liquid level in the
reservoir. The reason for this is that upon startup and initial operation of
the electrolyzer, the electrolyte can build up bubbles and foam rising in the
gas outlet tubes. If the foam rises over the high point of the tubing and
drips down towards the bubbler it can be released by opening the valve on
the first bubbler to let the KOH solution flush out. Be very careful if you do
this, as pressure is built up in the system and the caustic will spray out.
Wear protective clothing and face shield. Be sure the container you are
emptying into surrounds the nozzle of the valve. You can also attach a
length of hose to the outlet of the valve and then release the KOH to flow
down a small length of tubing aimed away from you and into the KOH con-
tainer. The length of the exit port tube should be a little over twice as long
as the hydrostatic tube, so that it can arc back down and connect with the
bubblers.
124
Electrolyzers
Connect to the power source
Make all your electrical connections to the electrolyzer from the panels.
Make sure they are tight so they will not slide. Cover the terminals of the
electrolyzer with liquid electrical tape and/or dip rubber compound or
silicone caulk.
Fill the electrolyzer
Finally, fill the electrolyzer by pouring KOH into the reservoir. Observe
all safety precautions, as mentioned above, for handling KOH. Let the
electrolyte fill up the column until half of the reservoir is filled. Put the
reservoir cover on. When this is set up, connect the exit port tubing to the
bubbler in the gas processing system.
As you like it
The P41 can be made any size you like, and with many variations.
Construction materials, method of construction and dimensions, can all be
changed to suit your preferences, needs and experimental design ideas.
The electrodes can be made larger or smaller, or configured into different
125
Electrolyzers
shapes. A large bank of these electrolyzers can be built at a very low cost,
which was one of our main goals for this design. You may want to build the
tanks from the regular schedule 40 PVC pipe which is more sturdy and
resistant to temperature change than the thin wall PVC. Of course, many
other materials can be used.
Although we used two contacts for the positive terminal contacts for the
electrolyzer, four could be used instead, to equalize the potential more
evenly around the mesh electrode. Four equipotential contacts may or may
not improve performance. We did not try this idea for this project, but it
may be worth investigating. Of course, adding more contacts also adds
more resistance, and if you try this you might want to gold plate the screws
and nuts to reduce resistance.
A bank of electrolyzers can be set up as you would set up a bank of bat-
teries, or solar panels, in a series, or parallel, or series/parallel connec-
tion; or, each electrolyzer can have its own dedicated power source. This
last option is actually the most energy efficient configuration.
The configuration you choose depends on a number of factors. If you
already have BSPMs on hand and wish to use them in a solar hydrogen fuel
cell system, its best to set up a bank of electrolyzers connected in series.
126
Electrolyzers
Quick comparisons
We did run a quick comparison test with another electrolyzer. This par-
ticular electrolyzer weighed in at about 30 pounds, used sintered nickel
plates, and was about three times as large as our P41. The P41 weighs a
little more than a pound and was less than half the size of the commercial
electrolyzer.
We connected the electrolyzers to matching solar panels and watched
the results. The P41 began gassing immediately, whereas the commercial
electrolyzer took about a half hour to release its contents into the feed
tubes. This fact pointed out design flaws in the commercial electrolyzer.
Basically, in any electrolyzer you want the gas to get out of the reactor
tank and away from the electrodes as fast as possible so that the gas does
not interfere with the process. It was obvious at first glance that the com-
mercial electrolyzer had gas pockets that had to be filled before the gas
would be released. Even after the other electrolyzer was given time to
come up to speed, it was visually evident that there was no contest.
The P41 produced twice the gas that the commercial electrolyzer pro-
duced. We also noticed that the P41 performed extremely well under inter-
mittent cloud cover as we had expected. Very sharp peaks and troughs in
127
Electrolyzers
gas production were evident in the commercial electrolyzer, whereas with
the P41, gas production was smoother and more consistent. Although we
cannot say scientifically that this was a direct result of the thermal fly-
wheel design, the comparison with the conventional electrolyzer demon-
strated the anticipated results.
Sharp peaks and troughs were evident in the other electrolyzer with
minor atmospheric hindrances that vary from second to second and or
minute to minute, such as moisture clouds or dust clouds. These are not
perceptible to the naked eye, but never the less affect the power output of
the PV panels and thus the gas production from the electrolyzer. Under
these conditions, the P41 exhibited a rolling effect with a more consistent
gas output, and with full cloud cover it was producing much more gas than
the commercial electrolyzer.
BSPMs and electrolyzers
BSPMs (Battery Specific Photovoltaic Modules) give more voltage than
current  the average solar panel delivers between 2 amps to 10 amps
short circuit current per 16 to 18 open circuit voltage. BSPM panels vary
in output but essentially are designed to provide voltage that is above 12
volts in order to charge 12 volt batteries. The voltage of a charging source
must be higher than the battery charged in order for charging to occur.
128
Electrolyzers
Although such panels are specifically
designed for charging batteries, they can be
used in systems where a bank of electrolyz-
ers is connected in series. In other words, if
you have solar panels that deliver a nominal
Two 12V 10 amp BSPMs
real working voltage of around 15 volts at 10
connected in parallel,
amps, you can power about three elec-
powering three 4V 20 amp
trolyzers with two panels, if you want to
electrolyzers connected
in series.
input 4 volts into each electrolyzer cell at 20
amps (see illustration at right).
Series connected electrolyzers
In a series electrolyzer connection, to determine what is needed for
solar panel output voltage to power your electrolyzers, add up the voltage
that each electrolyzer requires. If you wish to input more current than one
panel can supply, connect the two solar panels in parallel, which, if using
the same rated panels as just mentioned, will deliver 15 volts at 20 amps.
The advantages of using a series connected electrolyzer bank is that
you can use a common water or electrolyte reservoir, and BSPMs are
readily available to purchase. The downside of the bank configuration is
129
Electrolyzers
that if one of the electrolyzers
goes out, all the others go out;
and, pulling one out of the system
requires that the whole system
stops gas production, while the
Electrolyte connections for a
problem is being solved. Also if
bank of six series connected
one electrolyzer is performing
electrolyzers with a common
poorly it can take the other units
electrolyte reservoir.
down in gas production. And, with
common electrolyte, you have to deal with a larger amount of caustic in
situations where the electrolyte has to be drained. Series connected
electrolyzers act just like some Christmas tree lights, in that, when one
unit goes out, the rest of them go out. Generally, series connected elec-
trolyzers are not as efficient as parallel connected electrolyzers.
Setting up electrolyzer banks
If you use BSPMs with a common electrolyte reservoir configuration,
arrange the electrolyte feed tubing and electrical connections to allow
the DC current to proceed through each cell without diverting through
the electrolyte. To avoid diversion of current through the electrolyte, pro-
vide a path of greater resistance through the electrolyte since the current
130
will follow the path of least resistance. For example, for four 12 volt 10
Electrolyzers
ampere BSPMs connected in parallel-
series (that is, two pairs of parallel
connected panels, with the pairs con-
nected to each other in series), to pro-
vide 4 volts, 20 amperes to each of six
series connected electrolyzers, you
would set up your electrical connec-
tions and electrolyte tubing connec-
tions similar to the illustrations on this
A bank of six series connected
and the preceding page. You would also
electrolyzers with a common
electrolyte reservoir, powered by
two pairs of parallel connected
BSPMs that are connected,
pair to pair, in series.
have to make sure that the elec-
trolyte in the separate gas exit
tubes can not connect (as shown at
left). This would be controlled by
having the individual gas exit tubes
sufficiently high.
131
Configuration for gas exit tubes
for a bank of six electrolyzers.
Electrolyzers
Parallel connected electrolyzers
Parallel or unipolar connected electrolyzers are connected as shown at left.
This type of connection can not be generally used with BSPMs because they
would provide too much voltage for each electrolyzer cell. The advantage of
parallel connected electrolyzers is that they are more energy efficient and the
performance of an individual unit does not affect the others. If one goes down,
it can be replaced or repaired while the others are still operating and producing
gas. Each electrolyzer in the parallel connection has its own electrolyte reser-
voir. This is an advantage if the electrolyte has to be drained from the unit
because you will be dealing with smaller quantities of caustic material.
A bank of parallel connected electrolyzers requires ESPMs that can deliver
lower voltage and higher currents. For instance, to power 3 electrolyzers at 4
volts and 20 amps, you would have
to supply 4 volts at 60 amps to
power all three electrolyzers.
Operating with ESPMs that pro-
duce 4 volts at 20 amps per panel,
would require three panels con-
nected in parallel to produce the 60
Three 4V 20 amp ESPMs connected in
amps needed to power the three
132
parallel, powering three 4V 20 amp
electrolyzers.
electrolyzers connected in parallel.
Electrolyzers
Stand alone configuration
A stand alone is a configuration of one dedicated power
supply for each electrolyzer. Each electrolyzer, and its
power supply and water/electrolyte reservoir is isolated
One 4V 20 amp
from any other compo-
ESPM powering
nents in the system,
one 4V 20 amp
except for the gas exit electrolyzer
pipes which merge into a
common storage device. This configuration
requires the use of ESPMs. Thus, for an
electrolyzer that you wish to input power
Two 2V 20 amp ESPMs
connected in series,
at 4 volts at 20 amps, you would use an
powering one
ESPM that produces that exact voltage
4V 20 amp electrolyzer
and current.
Large electrolyzers vs. small electrolyzer banks
In designing a solar hydrogen system, you will need to determine
whether it will be more efficient and cost effective to use a bank of smaller
size electrolyzers, or a larger electrolyzer or two.
133
Electrolyzers
It is usually more cost effective and energy efficient to use larger elec-
trolyzers, but this is not always the case. A larger electrolyzer can easily
be built by simply scaling up the dimensions of the P41.
In scaling up, it is important to increase the current capacity in the con-
ductors and connectors. Bus and connecting wires and connectors have to
be larger. Larger size conductors and connectors are necessary so that
the electrical energy is not dissipated as heat. Wires and connectors can
get quite hot when carrying large amounts of current. Also, since you are
making hydrogen, it is wise to use amply oversized conductors. Make sure
that all terminals and connectors are rated for the current you will be oper-
ating with. Similarly, you can downsize this design for micro applications
and/or educational demonstrators.
Designing your own electrolyzers
It is easy enough to build an electrolyzer, but to design and build a very
good one takes time and consideration. For instance, there are many
types of materials that can be used for electrodes.
Some of the more exotic electrode materials are nickel coated with
manganese, tungsten or ruthenium oxides for positive electrodes. These
metals give quicker action for the part of the reaction that occurs at the
134
positive electrode. Nickel plated platinum can be used on the negative
Electrolyzers
electrode to increase the rate of hydrogen production. Also, gold or plat-
inum plated nickel can be used for both electrodes, and/or plain nickel or
nickel oxides. Raney type metals with their larger surface areas can also
be used. Monel, as noted earlier, makes a very good electrode and is rel-
atively inexpensive. There are also materials out there that have never
been used for electrolyzer electrodes just waiting to be discovered. The
shape and configuration of the electrodes and the electrolyzer vessel is
another area for exploration.
Electrolyzer designs historically have emerged from particular indus-
tries. Their design parameters and requirements are quite different than
what is needed to produce hydrogen from renewable energy sources. That
said, there is a lot that can be learned from electrolyzer development over
the years, but there is plenty of room for electrolyzer innovation, especial-
ly for those designed to operate with renewable energy power sources.
Keep in mind also, the highest performance electrolyzer may not be the
best for your application. Very high performance electrolyzers are gener-
ally more expensive to make and maintain, and the materials are not as
readily available. For instance, a Ferrari is a very nice high performance
car, however if one simply needs to drive to town everyday to buy the gro-
ceries, the Ferrari would be overkill. Many people make the mistake of
135
Electrolyzers
thinking that the highest performance electrolyzer is the best for their
application. Lower cost, lower performance materials can usually do the
job just as well. Reasonable costs for reasonable results is a good goal.
Electrolyzer performance testing
Alkaline electrolyzers usually operate between 1.6 volts and 2.3 volts,
with the average being around 1.8 volts, and at around 20 amps current
input. This is an electrical efficiency of between 54% to 78%. High pres-
sure and high temperature (which means more expensive to maintain)
electrolyzers can operate between 1.3 volts to 1.7 volts, with electrical
efficiencies of 95% to 73%. However the lower voltage does not allow
more current input. This means less gas production than the low pressure
alkaline system, even though the process is more  electrically efficient.
There is a relationship between current density and voltage requirements.
More voltage is needed to drive the circuit at higher current densities.
To generally test the electrical efficiency of an electrolyzer, take a cur-
rent reading and a voltage reading while the power supply is connected to
the electrolyzer and it is in operation. To take a voltage reading, apply mul-
timeter or voltmeter probes to the positive and negative electrode con-
nectors. This reading is the voltage draw of the electrolyzer. To take the
136
current reading in amperes, connect one probe of the multimeter or
Electrolyzers
ammeter to the positive connection of the power source and the other
probe to the positive lead of the electrolyzer. The current flows through the
meter and the reading will give the amps being drawn at the moment.
Multiply the current reading times the voltage reading to calculate the
watts being used by the electrolyzer at the moment of the reading. Divide
1.24 by the voltage reading to get the electrical efficiency of the elec-
trolyzer at the moment of the reading. For instance, if you get a reading of
1.8 volts at 16 amps input, then the electrolyzer at the moment is operat-
ing at 28.8 watts. Divide 1.24 by 1.8 and you will see that the electrolyzer
is operating at about 69% electrical efficiency.
Collect performance data for RE power sources
With most renewable energy power sources, voltage and current
changes constantly due to minor or major variations of light intensity, or
wind speed, etc., to name just a few variations depending on the power
source. Field testing and collecting the data over time is important to get
average output figures for real time situations. Average them out to figure
total gas production for a season.
137
Electrolyzers
Testing equipment
These field tests can be monitored and information stored on a com-
puter with inexpensive multimeters and the software that comes with
them. Most electronic stores have multimeters with software available.
These multimeters are quite versatile, but be sure to get one that will han-
dle your amp output and provide current readings. You can also use data
loggers to store information and then download it at periodic intervals.
For bench testing the general performance of an electrolyzer, construct
a power supply controller. This will allow you to vary the current and volt-
age input to the electrolyzer and test its performance and characteristics.
You can also measure and compare exact input in terms of current and
voltage, and the output in terms of volume of gas and so on.
Measuring gas output
Gas volume, flow and pressure can be monitored in many ways.
Manometers and flow indicators can be used, or the simplest method of
gathering gas statistics is to invert a three gallon bucket in water in a five
gallon bucket to create a floating water/gas storage tank. Although this is
not as accurate as other means, it will give you an idea of how much gas
is produced over a given periods of time with given voltage and current.
138
Electrolyzers
Gas production formulas
There are formulas to calculate gas production output under ideal condi-
tions. These calculations are handy to make comparisons with your field and
bench tests. For instance, to calculate the output of hydrogen and oxygen dur-
ing an hour, first, measure the voltage at the electrolyzer terminals during
operation. Then measure the amps as indicated earlier. Then, multiply the
voltage times the amperage to get the power draw of the electrolyzer in watts.
For instance, if the electrolyzer draws 2 volts at 20 amps, this would be
40 watts of power being used by the electrolyzer. Next figure out how many
joules are being used per second . Basically 1 watt per second equals 1
joule per second. 40 watts of power usage would be equivalent to 40 joules
per second. There are 3600 seconds in an hour, so multiply 40 times 3600
seconds. This equals 144000 joules per hour (144 kj).
One liter of water yields 1,358.3 liters of hydrogen and 679.15 liters of
oxygen. It takes 13,170.9 kj to disassociate one liter of water. So we divide
13,170.9 kj by 144 kj which indicates that it would take 91.46 hours to elec-
trolyze 1 liter of water and produce 1,358.3 liters of hydrogen and 679.15
liters of oxygen.
139
Electrolyzers
To find out how many liters per hour of hydrogen will be produced, divide
1,358.3 liters by 91.46 hours. The result is the liters per hour that will be
produced, and in this case would be 14.85 liters of hydrogen per hour.
Similarly, to find out how many liters per hour of oxygen will be pro-
duced, divide 679.15 liters by 91.46 hours. The result is the liters per hour
that will be produced, in this case, 7.42 liters of oxygen per hour.
To work with cubic feet of gas rather than liters, divide by 28.317. For
instance, 14.85 liters of hydrogen divided by 28.317 equals 0.524 cubic foot
per hour or about one half cubic foot per hour. And, 7.42 liters of oxygen
divided by 28.317 equals 0.26 cubic foot per hour. As you can see oxygen
production will always be one half of the amount of hydrogen production.
Please note that these volumes are calculated for what is termed SLC
(standard laboratory conditions) which is considered to be 24.47 liters at
25°C or 298°K and at a pressure which is one atmosphere or 101.3kPa.
You can also calculate for what is called STP (standard temperature and
pressure). This is considered to be 22.4 liters at 0°C (273°K) and
101.3kPa (one atmosphere).
140
Electrolyzers
The bench tests can be correlated with field readings and the above cal-
culations to get a rough idea of how much gas is generated during the
testing period, and how much storage space will be needed for the gas
within certain pressure parameters and so on.
Bench test results can be correlated with given characteristics of the
microclimate for any place. Daily insolation and/or wind speed conver-
sions to power output with given equipment parameters can be used to
ballpark the general output for a given location.
For a photovoltaic system, daily insolation tables are available quite
readily on the web or in various publications. For wind power there are
wind tables, but exact topographical characteristics are critical to per-
formance and general wind statistics will not usually suffice, so a site test
is really necessary.
141
Gas Processing System
A gas processing system prepares the generated gas for end use.
There are several devices involved, and a variety of possible configura-
tions. To design an appropriate gas processing system you need to have
a working knowledge of the type of gas or gases being processed. Here
we are concerned with hydrogen and oxygen gas
Hydrogen history and characteristics
Hydrogen derives its name is from the Greek hydro, meaning water, and
genes, meaning forming, thus  water forming. Hydrogen was recognized
as a distinct substance by Cavendish in 1776 and was later given the
name hydrogen by Lavoisier who noticed that water was formed when
hydrogen was burned.
Hydrogen is the most abundant element in the universe. Over 90% of all
the atoms, and thus about three quarters of the mass of the universe, is
hydrogen. This is one very simple reason why the planet will definitely
move into a hydrogen economy.
Although it is present in the atmosphere, hydrogen is not exactly a "free
floater," as it is chemically very active. It combines readily with other ele-
ments and, so is locked into compounds. On this planet most hydrogen is
142
Gas processing system
locked into water and organic compounds which make up about 70% of
the earth's surface. In the atmosphere it is present at only about 1 ppm
(part per million).
Hydrogen is the lightest of all gases and disperses quickly if not con-
fined. It is colorless, tasteless, odorless, and slightly soluble in water.
Hydrogen can be liquified at -423°F, and can take on a metallic state
under certain conditions. At about 120.7 kilajoules per gram, it has the
highest energy content of any known fuel. Its atomic number is 1, its atom-
ic symbol is H, and its atomic weight is 1.0079. Apart from the isotopes of
hydrogen (protium, deuterium, and tritium), hydrogen occurs under normal
conditions in two forms or kinds of molecules. These two forms are known
as ortho- and para-hydrogen. They differ from one another by the spins of
their electrons and nuclei.
Hydrogen can be produced by many different methods, most notably:
steam reforming, electrolysis, ammonia dissociation, and partial oxida-
tion. It can be stored for later use, as a gas, liquid or in compounds such
as hydrides. It is highly flammable and explosive and can be easily ignit-
ed through static electric discharge; or by a catalyst such as platinum in
air or oxygen without any other source of ignition.
143
Gas processing system
Hydrogen safety
Proper precautions and safety measures recommended in the MSDS
should be followed as well as other ruling jurisdiction safety rules and
guidelines when handling hydrogen. Please go the Resources page, follow
the links to the MSDS recommendations, ...and read them carefully!
Grounding
When working with hydrogen, make sure all metal components in the system
are grounded. This will bleed off any accumulating electrostatic charges that
can build up on a metal surface that is insulated to the ground. If the parts are
not grounded, your body will be the preferred path to ground in case of electro-
static buildup. One spark jumping from a component in the system can ignite
any leaking hydrogen in air. In a very humid area, this is not as much of a prob-
lem, but in a dry climate, take as many precautions as possible.
Ground all ungrounded metal components no matter what the climate.
Electrostatic charges accumulate on insulated conductive surfaces in many
ways. Snow or sand or dust blowing across a conductor can cause static
buildup. Exiting a vehicle can charge your body with static, which can be a
source of ignition when approaching a hydrogen system. Never exit a vehicle
and proceed to a hydrogen installation without first contacting a grounded
144
Gas processing system
touch plate. Whether exiting a vehicle or not,
you, and others approaching the hydrogen area
should always ground yourselves with a touch
plate before proceeding into the area.
Hydrogen proof seals
Hydrogen easily diffuses through the small-
est cracks and spaces, and it is important to
perform hydrostatic tests on as many of the
components and connections as possible; and
Grounded touch plate
to use pipe dope and tape that is compatible
beside the entry to a
with hydrogen. Teflon® tape is color coded for gas processing area
specific applications. Yellow tape is the color
code for gas line use, and would be the choice for hydrogen, although other
color coded tape could be used. For pipe dope, either Lox 8® paste or Super
Lox 8® is hydrogen compatible
Restrict access to the hydrogen area
Restrict access to the experimental area to only yourself and others
who understand the safety guidelines, rules and regulations that apply.
The hydrogen work area should be outdoors, and/or in a structure that is
145
Gas processing system
fully ventilated according to guidelines. It should be a
safe distance from all sources of ignition. It should also
be protected from unauthorized entry. Appropriate signs
should be placed according to guidelines to indicate the
material and nature of the hazard area and that no
source of ignition is allowed.
In other words you don t want your Uncle Bob coming over for a visit with
a cigar in his mouth asking perhaps for the last time,  What s this...? and,
yes, it can happen to you  not to mention curious kids or pets and wild ani-
mals. A serious experimental station should have a cyclone fence with
locked gate around the work area to keep such problems at a minimum.
Make electrical devices hydrogen safe
Any source of ignition should be strictly prohibited within your work
area. Electrical connections need to be appropriate for working in a hydro-
gen environment and connectors should always be tightened where they
will not move or be loose and cause a spark. Light and other types of
switches should be rated for the hazard of the environment. Battery oper-
ated devices with switches such as flashlights can also be a hazard.
146
Gas processing system
Hydrogen compared to other fuels
Because hydrogen is so light, it dissipates quite rapidly upward, and
this is a good thing. Other types of fuel gases are heavier and tend to
linger at ground level longer making them more dangerous. On the other
hand, hydrogen has a lower flammable limit of only 4% in air and a high-
er flammable limit of 75%. This gives it a much wider flammable and explo-
sive limit than other types of fuels.
Hydrogen in the presence of oxygen
Oxygen rich atmospheres also require more caution, as when working
with oxygen hydrogen fuel cells. Flashback arrestors should be placed
appropriately in your system so that any flame can be quenched before
proceeding to other parts of the system. Flashback arrestors need to be
placed between the storage tank and inlet into the fuel cell, and before
gas entry in the storage tank also.
Any catalytic material such as platinum, which can cause combustion of
hydrogen in air or oxygen rich environments, needs to be preceded by a
flashback arrestor. If you install a catalytic recombiner in the system, pre-
cede and follow the combiner with flashback arrestors.
147
Gas processing system
Connections need to be secure to avoid gas leaks and loss of pressure in
the system. If you are working with experimental systems such as outlined in
this book, you will be changing connections and reconfiguring them more fre-
quently, so that a certain flexibility needs to be built into the system. With
greater flexibility, comes the potential for much more leakage. Be aware of
this, and adjust accordingly by adhering to all relevant precautions.
Oxygen
The name oxygen is derived from the Greek oxys, sharp, acid; and
genes, meaning forming, thus "acid former." Priestley is generally credited
with discovering it. Oxygen's atomic number is 8, its symbol is O, and its
atomic weight is 15.9994. It is slightly soluble in water and becomes a liq-
uid at -297°F. It has nine isotopes.
Oxygen is about 21% of the earth's atmosphere by volume, and over 49%
of the earth's crust. It is colorless, odorless, and tasteless. It reacts with all
elements except inert gases and it forms compounds called oxides.
Although oxygen is inflammable it vigorously supports combustion with
materials that are flammable.
It is used in many industries for a variety of purposes. It can be produced by
electrolysis, by heating potassium chlorate with a manganese dioxide, or by
148
Gas processing system
fractional distillation of liquid air. It is non-toxic, and as a gas poses no haz-
ards except for its vigorous support of combustion with flammable materials.
Oxygen and safety
Because of its support of combustion, it is important to keep oxygen sep-
arated from hydrogen both in the production of this gas in the electrolyzer,
and at any other stage of gas processing and storage. Also, store oxygen
away from oils and greases and other hydrocarbons. Storage of oxygen
should be at least 20' from hydrogen tanks, or separated by a barrier at
least 5' high and rated for a fire resistance of at least 1/ 2 hour.
For connections, green color coded Teflon® tape is compatible with oxy-
gen and LOX-8®, Super LOX-8®, or Oxytite® are recommended pipe dopes.
Generated or ambient oxygen?
In your experimental system you may use the oxygen generated, or not.
Hydrogen oxygen fuel cells require both gases, but hydrogen air fuel cells
will not need generated oxygen because they get it from the ambient air.
149
Gas processing system
Moisture and fuel cells
The end use defines the quality and quantity of the gas needed. For our
purposes, the end use for the hydrogen generated is a fuel for fuel cells.
Most commercial hydrogen generating units incorporate a drying mecha-
nism that removes much of the moisture from the gas. This is not appro-
priate for a fuel cell system. Free water will be abundant and should be
removed by a water filter or series of water filters, but it is not necessary
to remove very fine aerosols by coalescer or to use water absorption tech-
niques. The reason is that the hydrogen side of the fuel cell membrane
needs to remain hydrated (moist at a certain level) to aid and maintain
proton transport for the operation of the fuel cell.
Also, in terms of moisture for the hydrogen side, consider the number of
stacks in the system. If there are a large number of stacks, there are more
membranes to keep moist. For one stack, less moisture is needed. Design
and develop water management according to your needs.
If you supply oxygen from an electrolyzer to a hydrogen-oxygen fuel
cell, be sure to remove water from the oxygen. The buildup of water on the
oxygen side of the fuel cell is more detrimental, because water is being
formed there by the action of the fuel cell. It can drown the membrane and
150
Gas processing system
make it temporarily ineffective and inoperative. On the oxygen side of a
gas processing system, you may wish to add an extra filter or coalescer
and also a drier.
Removing contaminants
It is important to remove and/or neutralize contaminants from the gas
generating system. In this particular system, potassium hydroxide is the
electrolyte. It can be removed by passing the gases through a scrubber-
bubbler filled with water and/or vinegar. Vinegar neutralizes the KOH.
Other particulates are also picked up by the scrubber.
Scrubbers and diffusers
Liquid scrubbers remove contaminants quite effectively. They can be made
even more effective by breaking up the gas into smaller bubbles with screens
or diffusor/air-stones. These devices expose a greater surface area of the
gas to the liquid removal agent and thus make the scrubbing more efficient.
Use a larger pore diffuser for this  finer pores take more pressure and clog
up faster. Better quality glass bonded silica diffusers have better resistance
to KOH and do not deteriorate as quickly as other types.
151
Gas processing system
Another option is to use polyethylene diffusers, and/or stainless steel or
monel metal screen diffusors. There are many viable approaches. If the
pores are too fine, more pressure will be needed and the diffuser will clog
more easily. Our system has a regular fiberglass hardware screen, which
makes a larger size bubble and thus scrubs less surface; but it does not
require much pressure to push the gas through.
Filters and coalescers
On the hydrogen side of
the gas processing system
there is no need to remove
all the moisture because the
membrane should be
hydrated. So, we use one
low cost water filter which
does the job adequately. If
you find that one filter is not
sufficient, add another.
Water filter (left) and regulator (right)
152
Gas processing system
A coalescer and/or drier is not necessary, however add one if you want,
based on the amount of aerosols in your particular system. There is no
end use for the oxygen in our system, but if you are going to use the gen-
erated oxygen, add a coalescer to the water filter to remove as much
water as possible. The number of filters and other devices, and their
size/capacity will depend on how many electrolyzers there are in the sys-
tem, and how large the electrolyzers are. In other words, it depends on the
amount of gas to be processed.
Recombiners
The solubility of both hydrogen and oxygen in water is minimal so it
does not readily diffuse and migrate. However, there is always some diffu-
sion of gases and mixing. If you feel a need to further purify your gas, you
may want a catalytic recombiner, which can be a stainless steel tube with
platinum coated alumina pellets inside it. The gas is passed through the
pellets, and water is formed and heat given off in the process. If you
include a catalytic recombiner in your system, put a water filter on the out-
put end and a flash arrestor on either side of the recombiner unit. There
are instruction for building a catalytic recombiner on page 174.
153
Gas processing system
Check valves and regulators
A check valve in the system prevents
backflow of gas when the system is not
operating. A regulator will provide the right
Check valve
amount of pressure for the fuel cell unit you
are feeding. The regulator can be configured
to deliver gas directly from the system, or from storage, or both. With this
system, a simple valve can be used because the pressure is very low and
is not enough to rupture fuel cell membranes.
Environment of operation
The environment in which your hydrogen and fuel cell system will oper-
ate is important to consider in the design of the components and the
choice of materials used to construct them. Consider temperature. In cli-
mates where significant seasonal changes occur, contraction and expan-
sion of components and the stress these changes will have on the system
should be considered when choosing the construction technique. For
instance, in one portable storage device we built, we used epoxy against
PVC. It was great for one season, but when the next season rolled around,
and we did our seasonal check we found that the epoxy delaminated from the
154
Gas processing system
surface. Cold weather contracted the materials and broke the epoxy bond.
Obviously, we needed to find other options for connecting these two sur-
faces for cold weather applications.
Even with the best and most expensive fittings rated for hydrogen use,
you can encounter problems with severe weather such as arctic conditions
or desert conditions, or extremely wet conditions and/or any seasonal or
daily changes  anything that creates stress or shows materials' incom-
patibility in rates of expansion and contraction.
System longevity
The fourth consideration is the longevity of the system. Is it experimen-
tal, with a short life expectancy, or should the system have a permanent
structure? The system components in this book are used and intended for
a short life experimental framework. The system is not designed for long
term use with hydrogen rated components. A long service system would
require more technologically mature and more expensive components.
If your experimental design performs to your satisfaction, you can then
upgrade to more durable components. High quality stainless steel compo-
nents should be used in every system intended for any length of service.
Other materials can be substituted, but they must stand up to the condi-
155
Gas processing system
tions required for the purpose and location of your system. Basically, most
natural gas fittings, pipes and connectors will suffice, but you must do your
homework and make sure they are rated for hydrogen use.
Gas processing system
The gas processing system for this particular solar hydrogen fuel cell
system consists of several gas scrubbers, a gas regulator, water filter, and
a few valves, check valve and flashback arrestors.
Gas scrubber
Gas scrubbers absorb contaminants from the gas that is fed through
them. Liquid gas scrubbers as we describe here, provide a safety feature
also, in that they act as flashback arrestors as well.
Liquid scrubbers, or bubblers as they are sometimes called, simply
release the gas through a liquid. The liquid absorbs contaminants, refining
the gas so that it will not be detrimental to the device being fed, in this
case, a fuel cell. We chose distilled water or vinegar, and a combination of
both, though there are other liquids that will work.
156
Gas processing system
Vinegar and distilled water
In this gas scrubber system the liquid absorbs particulate contaminants
and also removes any electrolyte aerosols that flow out of the electrolyz-
er with the gas. Distilled water is very effective at removing the electrolyte
from the gas, but the addition of vinegar also neutralizes the KOH, which
enhances the scrubbing process greatly.
If you use vinegar to scrub gas, make sure it is sulphur free. Natural
foods stores often have natural vinegar in bulk. This is your best bet for
getting vinegar with no additives. Sulfur compounds have a negative effect
on fuel cells, so be sure to avoid them. If you buy bottled vinegar, beware.
Some bottles state on the front that it is natural but if you read the fine
print, you will see that sulphur has been added. If you can, get red (wine)
vinegar. If you use red vinegar, as the vinegar gets depleted in its neu-
tralizing capacity, it turns a pale color which indicates that it is time to
change it. White vinegar works fine also, and is also cheaper to buy, which
can be important if you are going to use large quantities.
Our system has two scrubbers. Both can be filled with distilled water, or
one can be filled with vinegar and the other with water. If you decide to
use both vinegar and water, put the vinegar filled scrubber first in the sys-
157
Gas processing system
tem, followed by the water scrubber. This way the vinegar is also absorbed
from the aerosol stream in the second scrubber.
If you wish, use one scrubber filled with distilled water instead of two;
or you can have four scrubbers: four filled with water; or two filled with
vinegar, and two with water; or one with vinegar and three with water.
Experiment with different designs
If the scrubber has a large diameter, it will hold more liquid and will not
need the water or vinegar changed as often. There are many other ways
to construct bubblers. The main consideration is to get as much gas sur-
face area exposed to the scrubbing fluid as possible, without creating
pressure problems. The smaller the bubbles, the more surface area will be
scrubbed. Also, the further the gas bubbles have to travel through the
scrubbing fluid, the more they will be scrubbed. However, the height of the
liquid in the scrubber adds pressure that must be overcome by the gas in
the feed tube inside the scrubber, so don t make the liquid level too high.
System design largely consists of making compromises. I decided to
use larger mesh screen because I wanted to use the bubblers with a num-
ber of different electrolyzers, and I did not want to restrict the gas flow
unnecessarily. Also, since the bubblers would be sealed, The mesh cannot
158
Gas processing system
be accessed to be cleaned or replaced. With a finer
mesh, clogging problems could occur over time.
If I used silica sand air diffusors, this clogging
would definitely occur and the diffusors would have to
be cleaned with muriatic acid occasionally to remove
any particulate buildup
Air diffuser
Although I do not see this as an especially serious
problem, I still chose to make them as maintenance
free as possible, which meant that I would have to sacrifice some scrub-
bing action for longevity.
A finer mesh than the particular screen I used would work, and it would
have been preferable to use monel screen. However, I happened to have
fiberglass screen on hand, and decided to try it. You can go down to a fair-
ly fine mesh screen without clogging problems if you make sure that the
KOH solution is as particle free (filtered beforehand) as possible. Of
course if you design the bubblers to allow you to clean the screen, you can
used a very fine mesh.
159
Gas processing system
Making bubblers
Tool list
1
Taper pipe tap, high speed steel, bright finish /2  -14 NPT,
McMaster-Carr, part #2525A175.
1 45
Drill bit, round shank Hss, /2 " shank diameter, /64" size, 6" over-
all length, 3" flute length. McMaster-Carr, part #2933A57.
1
Value-Rite tap wrench, straight handle, style for /4 "-1" (6-25mm)
taps. McMaster-Carr, part #25605A79.
Value-Rite tap wrench, straight handle, style for 0-1/2 " (1.6-12.5mm)
taps. McMaster-Carr, part #25605A75.
Hacksaw, and other types of saws to cut metal and other materials.
Drill, either handheld or drill press.
Caulking gun.
Materials list for 2 bubblers
All should be available at local hardware or plumbing supply
stores, unless otherwise noted.
7
1 PVC nipple, threaded on both ends, /8 " OD, 3" length. Cut in half
for two 11/2 " pieces. Can use any other substitute that works.
Fiberglass screen, stainless mesh. Cut 2 pieces to size to cover
nipple ends.
2 pieces PVC pipe 3" ID 12" length.
160
4 #3 caps for 3" ID PVC pipe.
Gas processing system
Materials List for 2 bubblers (continued)
3
6 PVDF single-barbed tube fittings, 90 degree elbow x NPT male for /8 "
1
tube ID, /2 " NPT. McMaster-Carr, part #53055K191.
2 pieces 1" plastic tubing, 91/2 " long.
Clear silicone rubber caulking.
Other parts for gas processing system
1
1 Filter with zinc body and polycarbonate bowl, manual drain, /4 " pipe,
35 scfm max. McMaster-Carr part #4274K12.
3 3
1 PVC spring-loaded ball-check valve /8 " barb x /8 " barb,
Buna-N-Seal. McMaster-Carr, part #7933K33.
3 1
2 Barb hose connectors, /8 HB x /4 MPT, part of reducer unit connecting
from electrolyzer tube to 3 way valve, and one to fit into filter.
1 3
1 Threaded barb connector /2 HB X /8 MPT, barb tube reducer part, for
connection from electrolyzer to 3 way valve.
3 3
2 Easy-Grip PVC miniature ball valve straight, /8 "x /8 " Barb shut off
valves for KOH supply. McMaster-Carr part #4757K18.
3 3
2 Easy-Grip PVC miniature ball valve 3-way, /8 " barb x /8 " barb.
McMaster-Carr part #4757K58.
3 1
1 coupler, /8  x /4  FPT, for connection from electrolyzer tube to
3 way valve.
161
Plastic tubing, variety of sizes.
Gas processing system
Making a bubbler
This bubbler is constructed from a
12" length of 3" inner ID (inner diam-
eter) schedule 40 PVC pipe. Two #3
PVC caps are the bottom and top.
Tube connectors are PVDF single
barbed tube fitting, with a 90° elbow x
3 1
NPT male for / 8" tube ID, / 2" NPT.
Plastic 1" tubing, 91/ 2" long, extends
down into the bubbler tube from the
tube connector in the top cap.
Bubbler parts,
Fiberglass screen mounted on the
above. At right,
lower end of the tubing breaks up the
close-up of nipple
that will be
gas stream into smaller size bubbles.
mounted on the
bottom of the
Assembly
tubing inside
Drill and tap two holes across from the bubbler.
each other on one of the end caps.
This will be the top cap of the bubbler. The holes are the entry and exit ports for
the gas, and accommodate two 90° barb connectors. Drill and tap a hole in the
162
center of the bottom cap for the third barb connector. Trim the barb connector
Gas processing system
for the bottom cap, and one of the connectors for
the top cap, so that they will be flush with the inside
of the caps. One barb connector in the top cap
should be left untrimmed so that tubing can be
mounted on it inside the bubbler. Spread epoxy on
the threads in each hole and on each connecter,
then screw each connector into its port. Make sure
the outlets point in the right direction (see illustra-
tion on previous page). When they are properly
seated, to ensure a good seal, add epoxy inside
Looking up towards the
and outside the cap where the connectors meet the
top cap of the bubbler.
Notice gas exit port at
pipe cap surface. Let the epoxy dry for 24 hours.
about two o clock.
Cut a 91/2" long piece of 1" plastic tubing and
push one end over the piece of the barb connector that extends down from the top
pipe cap. Cut a small piece of fiberglass screen in a circle the size of the outside
diameter of the PVC nipple. Epoxy this to the free end of the nipple, then push the
nipple into the tube. You can also apply epoxy or silicone where the tube butts up
against the inside top cap to ensure a good seal. Let it dry for 24 hours.
The bubbler operates by conducting the gas through the connector, down
the tube, and into the liquid scrubber. The screen on the end of the inside
163
tube breaks up the gas into smaller bubbles. The bubbles rise to the surface
Gas processing system
and leave through the exit port. The bottom cap connector is attached to an
outside tube with a two way valve on the end. This tube is the fill/empty tube,
and liquid level indicator.
Coat the inside of the top cap side walls liberally with silicone. Then coat the
outside of the pipe edge with silicone for about 13/4 from the rim. Seat the
pipe cap and press all the way down. Put a bead of silicone around the rim of
the pipe and on the cap where the cap and pipe will be in contact. Smooth the
silicone out with your finger or other tool to get a good, clean seal. Do the
same with the bottom cap and let it dry for 24 hours.
Connecting the valves and tubing
Connect a 3 way valve to the left input port of the first bubbler with a 21/ 2"
3
piece of / 8" ID tubing (see photo next page). Connect another 3 way valve
3
to the output port of the second bubbler with a 21/ 2 long piece of / 8" ID
tubing. Connect the output port of the first bubbler to the inlet port of the
second bubbler with a 5" piece of 3/ 8" ID tubing. Connect a 21" or 22" piece
of 3/ 8" ID tubing to the fill/level ports on the bottom caps of the bubblers
Connect a piece of flexible tubing about 3" long to the top outlet of the 3 way
valve going to the left input port. Since the size of the tubing coming from the
electrolyzer is larger than the input to the bubbler, add a reducer. I used a 3/8 x
164
1
/4 FPT coupling, with a threaded barb connector, 1/2 HB x 3/8 MPT, and a 3/8
Gas processing system
1
HB x /4 MPT barb. This particular
combination is what was available at
the local hardware store, but you may
find a more suitable one piece connec-
3
tor reducer. This connects the /8 ID
tubing from the top of the 3 way valve to
1
the /2" ID tubing that goes to the
hydrogen gas outlet on the electrolyzer.
Connect a 61/ 2" piece of 3/ 8 tubing
to the outlet port on the second bub-
bler and connect that to the ball
check valve. Connect a 4" piece of
3
/ 8 tubing to the other side of the
check valve. This will be connected to
Two bubblers installed in
3 1
a / 8 HB x / 4 barb that has been
the gas processing train
inserted into the water filter. When
inserting this connector into the water filter, use Teflon® tape and pipe dope
to assure a decent seal.
Insert a 1/ 4 x 1/ 4 connector into the opposite port of the water filter, and
connect it to the regulator. Use Teflon® tape and or pipe dope on these
165
threaded connections. The regulator can be configured in different ways,
Gas processing system
The entire gas processing system
166
Gas processing system
and you can change the
configuration to suit your
needs. I simply used one
port for the inlet, and used
two ports for output with
1 3
/ 4 x / 8" barbs, one going
to the storage tank and the
other going to the fuel cell
Brass connectors and the regulator
stack for immediate use. I
1
closed the / 8" port on this
regulator with a cap, howev-
er, you could insert a pres-
sure gauge here if desired.
With this regulator arrange-
ment I can directly tap both
electrolyzer and gas stor-
age. Be sure to use Teflon®
tape and pipe dope on all
threaded fittings and make
sure they are secure.
167
Gas processing system
Remove the bowl from the water filter and apply Teflon® tape and/or use
pipe dope. If you do not do this, it will definitely leak gas, as the tolerances
are not precise on the threads.
The particular filter we used for this project has a polycarbonate bowl.
Polycarbonate is not well rated for use with potassium hydroxide. We
decided to use this however, as the system has a scrubbing system to
neutralize and absorb a good portion of the KOH solution, and the system
was not intended for long service. Polycarbonate does not stand up well in
harsh weather conditions with wide temperature fluctuations, and in gen-
eral tends to crack quite easily.
For long service, use a filter with a stainless steel bowl. These are more
expensive but necessary for a quality, trouble free system. Zinc bowls are
also available, but zinc is not compatible with KOH. Filter bowls also come
with either manual release or automatic release. With manual release you
have to drain the bowl on a regular basis by pushing on a plunger at the
bottom of the bowl. The automatic release relieves itself when it gets to a
certain fullness. In this project we used a manual release, but for a long
term system you might want to try an automatic release.
168
Gas processing system
Mounting the system
Using adjustable pipe hangers, attach the components to a backboard. I
used a 48"x 18" backboard for this purpose. Large size pipe hangers can
be purchased at local hardware, plumbing supply centers or home centers.
System mounted on display board
169
Gas processing system
Lay out your system configuration, attach the pipe hangers to the back-
board and then insert the electrolyzer, and scrubber units into the pipe
hangers. Adjust for tightness so that the units are held securely. Attach the
rest of the components to the backboard however you wish.
Component configuration
This system is designed to be a functioning display. Every part is in
view, and the parts and components can be easily changed and the
design modified during demonstrations so that the principles of the sys-
tem can be understood, observed, and worked on.
The electrolyzer and GPS system can be put together in a much small-
er space. The components can be arranged to give the system a very
small footprint. This particular configuration shows only the hydrogen side
of the GPS.
If you intend to use the oxygen generated by this system, you can add
bubblers, check valves, catalytic recombiners and whatever else you need
for the oxygen side. Since this project is for a hydrogen/air fuel cell, the
oxygen did not need to be processed and was shunted to a temporary
storage tank without scrubbing, or filtering. A check valve was placed on
the oxygen line to prevent back flow.
170
Gas processing system
Additions to the system
Gas detection system
Hydrogen gas detectors located in your work area can provide warnings
of any system leaks. In-line gas detectors, either hydrogen or oxygen, can
provide information on the purity of the gases in the system. The most
common gas detection technologies are electrochemical and catalytic.
It is important to understand the limitations, operating parameters, cal-
ibration, and maintenance for any detector that you use. URLs for detec-
tor information can be accessed from the Resources section. Most manu-
facturers have plenty of information about their detectors on their web-
sites. Neodym Systems offers low cost evaluation units, and they have a
variety of configurations that allow room for experimentation.
Another safety measure is to use flame detectors in case of a hydrogen
fire. Infra red detectors cannot detect hydrogen flames. You have to use an
ultraviolet flame detector that has an 1800 to 2500 angstrom range. This
range excludes the ultraviolet from the sun that reaches the earth, which
could cause false alarms. The 1800 to 2500 angstrom range UV that
comes from the sun is absorbed by the atmosphere. Hydrogen flames are
very hard to see  almost invisible  and people have actually walked into
171
Gas processing system
hydrogen flames because they could not see them burning. False alarms
can be triggered by lightening, corona discharges and welding being per-
formed at a distance. Detectors are great if they are used properly. If they
are not used properly they can be useless. Your AHJ (authority having juris-
diction) has the information about all the necessary safety precautions you
need to follow, and what they require for detection equipment.
Catalytic recombiners
You can further purify the hydrogen gas coming from the electrolyzer by
adding a catalytic recombiner. This will cause oxygen that has gotten
mixed in with the hydrogen to recombine with hydrogen into water mole-
cules. The water is then drained off through a filter or coalescer, which
creates a more oxygen-free hydrogen gas stream. Recombiners can be
put on both the hydrogen and oxygen side of the GPS system.
Safety considerations for recombiners
If you do add catalytic recombiners, put flashback arrestors on either
side of the catalytic recombiner tubes. In the recombination process, both
water and heat are created. The arrestors make it a safer process. It is
also necessary to provide a drier gas to the recombiner tube. Too much
moisture on the catalytic bead surfaces will impede the action.
172
Gas processing system
To provide a dryer gas to the tube, pre-
cede the recombiner tube with a coalescer
and then a dry type flashback arrestor, as
shown at right. A liquid bubbler type
arrestor would add moisture to the gas
stream, which is not desirable at this
point. A simple system with a coalescer
followed by a dry arrestor at the feed
point, and a filter/coalescer and bubbler
arrestor at the exit point is appropriate.
The use of a bubbler arrestor at the exit
point humidifies the hydrogen again which
makes it more suitable for fuel cell applications, as the hydrogen side of
the membrane needs to be kept moist for proton conductivity. However, if
you store the gas as metal hydrides, don't use a liquid bubbler on the exit
end. For metal hydride storage, you want a dry flashback arrestor to keep
the moisture level down. Moisture, oxygen and other contaminants will
impede the hydriding action, so keep the hydrogen gas as pure and dry as
possible. To use the hydrogen gas from the hydride bottle for fuel cells,
you will need to humidify it by moving it through a bubbler to the fuel cell.
Gas purifiers, arrestors, etc. can be purchased from companies such as
173
Harris Calorific, Inc.
Gas processing system
Building a recombiner
You can make your own recombiner
1
with 0.5% platinum on /8" alumina pel-
lets. These can be purchased, for
instance, from Alfa Aesar (part # 89106).
A catalyst container can be construct-
ed from a stainless steel tube with
screw threads on the outside at both
ends that will connect to threaded gas
connectors with barbed hose connec-
tions. The stainless steel tube should be
3
about six to eight inches long and / 4
to 1" wide inside diameter. Four silicone
or Teflon® o rings will be needed, two for
the top and two for the bottom inside the
tube. These hold stainless steel wire
mesh that hold the catalytic beads
inside the tube. The mesh should have
openings smaller than 1/ 8" to hold in the
1
/ 8" pellets, but the mesh openings
174
Catalytic recombiner
Gas processing system
should not be much smaller than what is necessary to retain the pellets, or
the gas flow will be impeded. The recombiner should be placed in a vertical
position, and the gas should enter the top and exit through the bottom.
Purging option
You can add an optional purge routine to this system by simply using bot-
tled nitrogen with a two stage regulator.
Purging the system with an inert gas removes the air from the system
before gas production. This gives a purer gas stream at the starting gate,
and is an added safety measure. Without a purge system routine, you must
run the system until the air is pushed out and replaced by the gases emit-
ted by the electrolyzer.
Several experimenters have noticed that some of the experimental PEMs
on the market recently are quite thin, and that if their systems or fuel cells are
not purged, they get what is called in the trade  membrane blowout. The oxy-
gen in the system air that was not purged out, and the hydrogen combine in
the presence of the catalyst on the membrane, and burn micro holes in the
cheaper thinner membranes, resulting in a  pop or system failure.
175
Gas processing system
I cannot recommend using MEAs with super thin membranes. It is proba-
bly best not to get anything thinner than a Nafion 112. Although I have heard
that some of these thinner membranes have been improved, I would be cau-
tious about using them. You might want to experiment with a one cell con-
figuration before investing and building stacks with these types of thinner
materials. Most of these membranes have not had a great success in the
market. The operating environments for fuel cells can be quite severe.
Although delicate components may work in the lab under ideal conditions,
they may not work well in the field. It should be noted that these membrane
failures may also be the result of not sealing the MEAs appropriately.
We do not use an inert gas purging system and have had no trouble to
date, but that does not mean it can t happen in the future.
In this system, you can purge with an inert gas by evacuating the sys-
tem (with a hand or motorized pump) and attaching a hose from one of the
stopcock valves, to a two stage regulated bottle of inert gas. Fill the entire
system at about two psi. A separate purge valve can be added to the sys-
tem. Bottled gas can be purchased from any gas supplier such as
Merriam-Graves and Pax Air.
System purging can also be done by running the electrolyzer and
immersing the tube that comes from the regulator, in a container of water.
176
The air will be pushed out of the system and be replaced by hydrogen
Gas processing system
completely up to the point where you are releasing the gas into the water
filled container. How long it takes for the hydrogen to replace the air will
vary depending on how much tubing you have in the system and how fast
the electrolyzer is releasing hydrogen, amongst other factors. The water
filled container also acts as a flashback arrester during this process. If you
are using the double 55 gallon storage tank for gas storage, at the top of
the lower storage drum there is a tube that connects to the regulator hose.
The bottom 55 gallon container is filled with water, and if you raise the
hose to just above the water level in the tank, the water level in the tube
will indicate the water level in the tank. Hold the water filled container with
the regulator hose in it up to the point where the tube is slightly above the
water level of the bottom drum. Immerse the tube coming from the drum
into the water container, and then lower it a little bit. Whatever air was in
the tube will be pushed out, and then you can connect both tubes under-
water with a double barb connector.
Pressure gauges, indicators, and switches
Simple pressure gauges and or electronic pressure gauges can be
included in the system for monitoring; as well as solenoid switches with
transmitters to turn gas flow on and off, change flow rates, and/or just sim-
ply read pressure. If you use any electronic switch or monitor it must
177
Gas processing system
conform to current safety standards for a flammable gas atmosphere. The
total gas production electrolyzer and GPS system can be wired to give
readouts for everything from current and voltage draw, to pressure and
flow of gas. The data can be transferred to a computer for data logging via
RF transmission. Infrared or visible light transmitters can also be used for
this purpose.
Component upgrading
After you have experimented and are ready to finalize a system, you can
upgrade to more durable hydrogen technology components. There are
many types of piping materials, flexible or rigid, and many types of fittings
available from a wide variety of vendors to suit your purposes. Generally,
natural gas fittings will suffice for hydrogen application. Be aware howev-
er, that brass fittings are not recommended for anything more than short
term experimental applications because the caustic atmosphere of KOH
deteriorates brass.
Pipe dope should be rated for oxygen use on the oxygen side of the sys-
tem, because some pipe dopes will deteriorate in the presence of oxygen.
A simple search on the internet will help you find the appropriate product.
178
Gas processing system
In certain engineering circles, pipe tape is not considered to be a
sealant, but a threading aid only. This is an important consideration when
dealing with hydrogen. The small size of the hydrogen atom means that it
will leak through any avenue available to it. Teflon® tape can be used with
natural gas fittings but the addition of pipe dope would improve the seal -
or simply use pipe dope without the tape. Pipe dope (rated for the service
involved) is the preferred sealing agent no matter what type of fittings you
are using.
Gas storage
Hydrogen and oxygen can be stored in a variety of ways: storage bags,
hollow tanks, or tanks with hydrides, or it can be stored in liquid form. There
are also methods being developed such as the use of carbon nano-tubes,
and metal organic frameworks, but these are not yet mature technologies.
Liquid phase storage
Liquid phase storage is compact, but it is not as economical as other
forms of storage because of the energy needed to lower the temperature of
the gases. It also requires a doubled wall Dewar type of container to hold
the liquid hydrogen and oxygen. However, recent advances in magnetic
refrigeration may, in the near future, make it viable to develop a renewable
179
energy liquification system for hydrogen. Large capacity Dewar containers
Gas processing system
are available on the surplus market for cryogenic storage of gases at rela-
tively reasonable prices.
The advantages of magnetic refrigeration are that it does not use envi-
ronmentally damaging chemicals, and, at present it is about 60% efficient,
compared to 40% efficiency for currently available gas compression refrig-
eration units that rely on a vapor cycle.
Hydride storage
Hydrides contain a relatively large quantity of gas in a small area. This
technology is compatible with renewable energy systems, but the draw-
back is the amount of energy necessary to maintain such a storage sys-
tem. Dedicated photovoltaic panels or wind generators would need to be
added to compress the gas. Hydride storage has a lot of promise for com-
pact hydrogen storage, and is a good avenue of inquiry for the renewable
energy enthusiast. An explanation of the hydride process and a practical
method of working with hydrides is discussed in The Solar Hydrogen
Chronicles, edited by Walt Pyle.
Metal organic frameworks
Currently, metal organic frameworks (MOFs) also look very promising for
180
solar hydrogen systems. These materials do not require high pressures and
Gas processing system
high temperatures to operate the system, so they do not require the level of
extra energy input that hydride systems need. With only modest pressure at
room temperatures, hydrogen can be adsorbed and retrieved in MOFs. One
gram of a metal organic framework can have a surface area of 3000 square
meters. It is also a very inexpensive material.
Low tech alternatives
Probably the most convenient gas storage options for the average
renewable energy experimenter are low pressure storage in bags; and low,
medium or high pressure storage in hollow tanks.
Containers for low pressure storage (under about 60 psi) can be bladder
tanks, pillow tanks, collapsible pressure vessels or LPG tanks. Most flexible
tanks store gases at just a little above atmospheric pressure. Manufacturers
such as Aero Tec Laboratories offer flexible storage devices.
LPG tanks or other tanks made with a good grade of stainless steel are
sufficient for medium pressure storage (60 to 125 psi). Stainless steel with
a high nickel content and very low carbon content is more resistant to
hydriding, otherwise known as embrittlement. (Hydriding is a great char-
acteristic for the storage medium in a hydride storage system, but it s not
a good characteristic for containment tanks!)
181
Gas processing system
Embrittlement occurs when the metal absorbs hydrogen which can
cause the metal to crack. If you make your own tanks, to prevent hydrogen
from getting into the metal, do not use water while welding. Also, preheat
the metal before welding and allow the weld to cool slowly. Embrittlement
proceeds faster at high pressures and higher temperatures.
High pressure storage, around 2000 psi and up
requires specially constructed high pressure cylinders.
Double drum storage
Short term storage can also be accomplished in
tanks or drums that would not be considered or used
for long term storage. For instance, I use recycled 55
gallon drums for oxygen and hydrogen storage. These
drums are very inexpensive and I can quickly weld the
tops on, fabricate a relief port and inlet/outlet port
quite easily. With some drums I have epoxied the tops
on. For me it works, as I do not use them for more than
one season. The drums are readily available and the
price is right. Two recycled
55 gallon drums
I do have to be concerned about what was stored in
for short term
182
gas storage
the drums beforehand. Contaminating the hydrogen
Gas processing system
supply with prior drum contents can be a problem. The drums must be
absolutely clean from contaminating substances. Although I use this tem-
porary storage method for my experiments, I must caution the reader that
storage in any thing other than your AHJ approved storage devices is not
an officially sanctioned practice and is not considered safe.
For drum storage, two 55 gallon drums are needed. One drum is securely
positioned above the other. The top drum has a constantly open vent which
releases any excess gas that cannot be stored in the drums. The bottom
Low pressure
storage in
double tanks
using
water pressure
183
Gas processing system
drum has a gas inlet near the top of the drum. It has an outlet on the other
side near the bottom. A transfer tube connects the bottom tank to the top
tank. The bottom tank is filled with water and when gas is routed to the
tank, it pushes the water through the tube and through to the top tank. The
gas is partially pressurized by the weight of the water in the top tank. It is
a simple affair and works quite well.
Any type of general gas or barb fittings can be used to make the gas
outlet port and inlet and transfer ports. I usually epoxy these into place.
The regulator needs to be positioned at a height above where the gas
feeds in to the bottom drum, so that water does not flow into the regulator
when the bottom drum is filled with water. I could expand my storage by
connecting up more pairs of drums.
Floating tank storage
A water tank storage device can be built, very much like those used to
store methane in some rural areas. This consists of an inverted drum or
tank within another tank or drum that is filled with water. As the gas fills
the inner tank, the inner tank floats upwards in the water, and is kept from
tilting by guide rails or guy wires. A stop keeps the tank from rising total-
ly out of the water. For a more permanent setup, round concrete spring
184
tiles can be put in the ground and a inverted tank inserted for larger area
Gas processing system
gas storage. This is an easily built
structure and relatively inexpensive,
the most costly part usually being the
fabrication of the inverted tank. Tanks
can be made from fiberglass, and
other resin cloth type products. Large
size tanks can also be purchased from
such places as Aquatic Eco Systems,
or other agricultural suppliers.
Calculating tank capacity
To calculate the cubic foot capacity
of a cylindrical tank, multiply 0.79
times the diameter. Multiply this by
Floating tank storage
the diameter again and then multiply
that result by the length. For instance,
for a drum or container 7 feet in diameter and 10.4 feet in length, your cal-
culation and result would be: 0.79 x 7 x 7 x 10.4 = 402.584 cubic feet. To
convert this to US gallons, multiply 7.5 x 402.584 (there are roughly 7.5 US
gallons per cubic foot). So the total gallon capacity for this tank is about
3019.38 US gallons.
185
Gas processing system
The above equation gives a ballpark figure for storage at atmospheric
pressure (14.75 psi). If you are considering a higher pressure storage
system, such as around 60 psi, you could store about four times as much
gas in the above example in the same size tank.
Adding pressure
If you want more pressure in a storage system than an electrolyzer
pressured system supplies, use a compressor such as a diaphragm type
that is rated for hydrogen use. Be sure that all tanks, hoses and fittings
are rated for the pressure desired. Compressors of this type can be pur-
chased from such suppliers as KNF Neuberger and others. These
diaphragm pumps keep gases contaminant and oil free, and usually have
Teflon® coated diaphragms and ryton heads, which allow them to operate
properly in the somewhat caustic atmosphere of aerosol potassium
hydroxide solution that is used in the electrolyzer for this system.
Safe storage
Storage of hydrogen should be in a protected area, preferably with a
fence around it, marked by signs indicating the presence of flammable
hydrogen, and signs saying that no smoking or ignition devices are allowed
in the area. Drums and tanks should be grounded to dissipate static elec-
186
tricity and carry off induced charges from nearby lighting strikes.
Gas processing system
Setup and check the system
1. Make sure all tubing is connected in the system and is secure. This
includes all tubing related to the electrolyzer as well as the gas
processing and storage components.
2. Open the valves on the bubblers and fill with vinegar and/or water
to the top of the bubblers. Close the valves after filling.
3. Check to see if the valves are turned appropriately to direct the
flow of gases where you want them to go. If you are going to purge,
do so, and then reset the valves for start up.
4. Fill the reservoir with electrolyte and make sure the electrical con-
nections are made and are secure to the solar power supply.
5. Turn on the power to the electrolyzer and be sure the regulator is
open to the storage and other lines.
6. Observe each component while operating to see if any leaks or
problems occur.
If everything checks out, you are ready to connect to the fuel cell stack.
187
Planar Fuel Cell Stack
Fuel cell basics
Simply put, a fuel cell is an energy conversion device. It has no moving
parts and thus operates silently. In the fuel cell process, energy is
released as heat and electricity.
The process is: hydrogen is fed to one catalyst electrode, which facili-
tates the separation of the hydrogen atoms into electrons and protons.
The protons or hydrogen ions move through the membrane toward the
other catalyst, which is fed with oxygen. The stripped electrons cannot
pass through the solid electrolyte membrane or liquid electrolyte, so they
must be routed through an external circuit. The external circuit contains an
electrical load such as a motor or light bulb, etc., and leads to the other
catalytic electrode, where the protons and electrons recombine and bond
with oxygen to create water molecules.
If you would like to see an animation of this process and read a thor-
ough treatment of the history and basic functioning of fuel cells, refer to
my book Build Your Own Fuel Cells.
188
Planar Fuel Cell Stack
Types of fuel cells
There are many types of fuel cells. The most common types are:
AFC alkaline electrolyte potassium hydroxide
PEMFC (proton exchange membrane) uses fluoropolymer or similar
type membranes such as SPEEK. Microbial and direct
methanol fuel cells fall into this category also.
PAFC electrolyte phosphoric acid
MCFC electrolyte molten carbonate
SOFC electrolyte solid oxide
Each of these fuel cells are named or defined by the electrolyte used in
them, that is alkaline, phosphoric acid, molten carbonate, solid oxide, pro-
ton exchange membrane, etc.
The only deviation from this pattern is the DMFC (direct methanol fuel
cell) which uses methanol as a fuel without intermediate reforming; and
microbial fuel cells that use sugar as a fuel and derive current from the
metabolic activity of yeast. Both types use a solid ion exchange membrane
type electrolyte (proton exchange membrane).
189
Planar Fuel Cell Stack
PEMFCs have a solid ion exchange membrane made of sulfonated flu-
oropolymer, or a sulfonated polyetheretherketone (SPEEK) which is the
electrolyte, and for the most part uses platinum catalysts. There are other
materials in use for combination type membranes. It should be mentioned
that currently SPEEK membranes do not hold up as well as the fluo-
ropolymers such as Nafion®, but research is ongoing to produce a more
reliable and longer lasting membrane.
At the present time, the Nogoya Institute in Japan is making progress
developing a new glass based electrolyte that is much less expensive than
fluoropolymer membranes, but just as durable. In the near future, these
membranes may replace Nafion® and SPEEK in PEM cells.
SOFCs have solid yttria stabilized zirconia as an electrolyte and
perovskites as a catalyst.
PAFCs have a liquid phosphoric acid as an electrolyte, and
platinum catalysts.
MCFCs have a liquid alkali carbonate mixture and nickel catalysts.
AFCs have a liquid potassium hydroxide catalyst and
platinum electrodes.
190
Planar Fuel Cell Stack
All of these fuel cells have different operating temperatures. For
instance, alkaline cells run from 50°-200°C, PEM cells run from 50°-
100°C, phosphoric acid run at about 220°C, molten carbonate run at about
650°C, and solid oxide run at about 500°-1000°C.
PEM fuel cell configurations
For the purpose of this book we decided to work with the proton
exchange membrane cell. Proton exchange membrane or PEM cells are
easy to work with, and come in a variety of shapes.
The most common configuration is the block type fuel cell stack. Other
configurations are tubular and planar (flat) stacks.
Different shapes serve different purposes. Block stacks are convenient
to use in some applications and planar stacks are easier in others For
instance, a laptop computer would do well to use a planar configuration
rather than a block configuration. Generally, fuel cells can come in any
imaginable shape and size, and be designed specifically for a wide vari-
ety of applications.
191
Planar Fuel Cell Stack
Planar fuel cell stacks
We decided to work with a planar design in order to compare the effi-
ciency of block stacks to planar stacks.
Block stacks, although convenient in some respects, have problems
with water retention and drainage. This can be overcome, but it is never
the less a complication. Block stacks can also require fans to force pull
air through larger stacks, and this adds a rather dumb note to energy
efficiency.
Our planar design does not rely on convection/air or pure oxygen feed
(see Build Your Own Fuel Cells for more about convection and oxy-
gen/hydrogen fuel cells). The planar design simply relies on ambient air
ports. The intent was to see if these ports would give greater air intake, as
well as more efficient water dispersion. We surmised that this design
would take care of these two problems, and reduce maintenance problems
(and thus cost) over a period of time.
As with the other components of this project, primary considerations
were cost, availability of parts, and reasonable construction methods that
do not require high tech tools.
192
Planar Fuel Cell Stack
The result of this endeavor was the L79 planar stack. Although we are
still experimenting and tweaking the design, we believe it is a good design
to present with the rest of the hydrogen system.
If you have never built or tinkered with fuel cells before, I suggest that
you first work with a more conventional fuel cell design, before attempting
to build a planar stack like the L79. My book Build Your Own Fuel Cells
will give you a good background in fuel cell design and construction so
that you will be better able to construct and/or design a planar stack.
193
Planar fuel cell stack
Build the L79 planar fuel cell stack
Tools list
Available from local hardware stores, unless otherwise noted.
Scotch Brite® Pads
Tray, tongs
Sand paper or fine file
Caulking gun
1
/8" router bit, Micro-Mark, part #60719
Drill press or milling machine
Multimeter electronics store
Laser printer, local print shop or copy center
Drill, either handheld or drill press
Hacksaw, and other types of saws to cut metal and other items
194
Planar fuel cell stack
Materials list
All should be available from local hardware stores, unless otherwise noted.
Press-n-Peel PCB transfer film, Techniks, Inc., #20PNPB, pack of 20 sheets,
PNP blue.
Clear silicone rubber caulking
Plating kits and supplies, Caswell Plating
Barb hose connectors, one 1/4", and one 1/2"
6 Screws, round head machine, stainless steel, 6-32x 11/2 ;
6 washers, 7/32 , stainless steel;
6 nuts, wing or otherwise, 6-32(1/8) USS, stainless steel
3 Aluminum channel bars, 12" long, 3/8"x 3/8".
Silicone rubber gasket, .020" thick, one 12"x 12" sheet, McMaster-Carr
part #86435K35.
12 Membrane electrode assemblies, Element 1 Power Systems. Raw
membranes, untested, are $20.00 each when you buy more than ten.
PVC Type 1 sheet, 12"x 12 , 1 thick, McMaster-Carr, part #8747K118. I
used 1/2" thick, but I recommend using the 1 thick sheet instead.
195
list continued on next page
Planar fuel cell stack
Materials list (continued)
2 Copper clad PC boards, one sided, 8"x 10", 1/16" thick. Ocean State
Electronics, #22-264, or other PC board supplier. Make sure the
dimensions are correct.
1 Silicone rubber sheet, 3/32" thick, 12"x 12", 40A durometer plain back.
McMaster-Carr part #8632K431
1# Ferric chloride etchant. Ocean State Electronics or other supplier.
Ocean State Electronics #ER-21.
Constructing the L79
The L79 is easy to build and uses a few common materials in a unique
way (see illustration, next page). It consists of a six layered sandwich
composed of one PVC end plate, two PC (printed circuit) board elec-
trode/gas flow field plates, one 12 PEM MEA (membrane electrode
assembly) layer, and 2 rubber gaskets, one of which also acts as a gas
supply line.
This design allows single stack units to be racked next to each other
and fed by a gas manifold. The design can easily be altered so that a
wider area stack can be constructed with more PEM cells on it, or with
196
PEM cells that have a larger surface area for more current output.
Planar Fuel Cell Stack
The component layers of the L79 fuel cell stack
197
Planar Fuel Cell Stack
Each cell in this planar stack is connected in series, that is, the positive
electrode of one cell is connected to the negative of the next cell and so
on. In a block stack this is accomplished by using bipolar plates, but in the
L79 we use circuit board traces with tab wire connects to perform the
same function as the bipolar plates. The L79 can also be built in a paral-
lel configuration or a series-parallel configuration depending on the cur-
rent and voltage desired.
Selecting the materials for the electrode/gas flow field
In the L79 design, both the gas flow fields and electrodes are made
from single sided copper clad circuit board. Copper clad circuit board
comes in a variety of sizes and is either clad with copper on one side only,
or both sides. Boards also differ in the type of base material they are
made of. This fuel cell requires the FR-4 glass epoxy resin base, clad with
copper on one side only. We used a 1 ounce, which is coated with copper
to a thickness of .0014".
The FR-4 was chosen because it is very stiff, which is necessary for this
design. It is important for the electrode grid structure to be as stiff as pos-
sible so that it maintains an even contact surface over the membrane sur-
faces within the assembly, especially when it is tightened with the pres-
198
sure adjustment screws.
Planar Fuel Cell Stack
You can experiment with other types of copper clad board with different
base compositions, just remember that stiffness and machinability are the
key concerns.
Preparing the electrodes for template transfer
The first step in fabricating the PC board electrodes is to clean each
board thoroughly. You can use a Scotch Brite® pad and water to clean the
circuit board. Even though the circuit board looks shiny and very clean,
rub it all over with the pad and then rinse with water.
After you clean the board, be sure not to touch the copper surface with
your fingers as this will leave minute traces of oils and dirt that can inter-
fere with the processes of image transfer, plating, and the electrical per-
formance of the surface. A good habit for working with copper clad is to
wear clean thin cotton gloves to protect the surface of the board while you
handle it.
Dry the surface of the board with a lint free cloth. Inspect the edges of
the circuit board for burrs. If there are any, file them off. The idea is to get
a clean smooth surface, so that the transfer film that will lie flat on the
copper surface. Be sure that after you clean and wash the board that no
residue is left on the surface.
199
Planar Fuel Cell Stack
Transferring the templates
The templates for the oxygen and hydrogen flow fields/electrodes in this
book (see page 248) are intended to be printed on plastic transfer film and
then transferred to copper circuit board for etching. This method is com-
monly used to transfer printed circuit board designs to the copper surface
of the boards. We used Press-n-Peel®, which is a plastic film that has an
emulsion on one side and comes in sheets of 81/ 2"x 11" size. The plastic
sheet is loaded into the printer just as any other 81/ 2"x 11" paper would be.
Use a laser printer to print the templates for the two circuit boards on the
film. Print the image on the dull (emulsion) side of the transfer film. If you
do not have a laser printer, you can print out a paper copy on whatever
type of printer you have and then get a laser copy at a copy shop.
If you use a laser printer, do not let it warm up before printing. Print the
template to the transfer paper immediately after turning on the laser print-
er. The reason is that the heat from a laser printer will affect the surface of
the transfer film and can produce defects in the image. It is important to
get a very sharp, clean, well covered, distinct image, as the transfer film
will pick up the slightest imperfections.
200
Planar Fuel Cell Stack
There are two circuit board templates, the oxy-
gen side template, and the hydrogen side tem-
plate. When you have finished printing your tem-
plates on the transfer film, lay the film emulsion
side down on the copper surface of each of the
circuit boards.
Above, the oxygen elec-
For Press-n-Peel®, heat from a clothes iron is
trode template printed
applied to transfer the template to the copper. The
on transfer film. Below,
lay the film emulsion
suggested starting temperature is 275° to 325° F,
side down on the circuit
which is usually the acrylic and polyester setting.
board, and secure the
Not all irons are the same, and you may have to
film to the circuit board.
experiment a little to get the right heat. I found that
using the higher "linen"
setting worked better for
these large sized boards.
You can attach the
transfer film to the board
with some adhesive dots
as I did to hold the film
in place on the edges
201
when you start to iron.
Planar Fuel Cell Stack
Place a piece of paper between the iron and
the transfer film to help the iron glide better
over the surface. Move the iron very slowly
over the surface area of the whole board mak-
ing sure to not miss the edges.
After about ten minutes of slowly moving
Use a piece of paper
the iron around on the surface, lift a bit of the
between the iron and
film up from the surface of the copper clad
the transfer film
and see if the image sticks to the surface of
the copper. If it is not sticking, or parts are missing, iron over those par-
ticular parts or go over the surface again slowly until the whole image is
transferred.
Keep checking the progress of the transfer by gently peeling the film
from the copper surface every once and a while to see if the image is
totally and satisfactorily transferred.
Please note that there must not be any continuity breaks in the copper
surface. This does not mean no blemishes or imperfections, but does
mean that if you put a multimeter probe on one side of a copper trace
there should be a continuity beep when the probe is placed on the other
side of the copper trace. Take your time with this step. Keep ironing until
202
Planar Fuel Cell Stack
all of the transfer is complete. This can be a
tedious operation but it is a very important to
do it well.
When the ironing is completed to your sat-
isfaction, take the circuit board and transfer
sheet, and run it under cold water. After it has
When ironing is complete,
cooled down, slowly peel the transfer sheet peel back the transfer film.
off. You should see a perfect representation of
the electrode grids in a blue color covering
Small imperfections
the parts that will remain copper clad after
can be corrected
with a marking pen
etching. Full directions come with each set of
transfer sheets.
If there are minor imperfections in the
transfer ink representations of the electrodes,
you can make corrections by filling in small
missed spots with a fine tipped Sharpie®
marking pen.
203
Planar Fuel Cell Stack
Etching the board
You will need a container to etch the board in. This can be anything from
a photo developing tray to a Rubbermaid® container. Do not use a metal
container, spoon or other utensil made of metal. It must be plastic or
glass. Do not use any containers that will later be used for food.
A small cat litter box works well and can be purchased from any hard-
ware or pet store for a few dollars. Get a size that will easily accommodate
the size of the circuit board with a little wiggle room, but not too much big-
ger, as that would require more etchant  which would be a waste just to
accommodate the oversized container.
When working with etchant solution, wear rubber gloves to keep the
chemicals from contacting and staining your skin. Also, wear old clothes.
The stain from the ferric chloride will ruin any clothing and the stain will
not come out. Make sure you have plenty of rags around to wipe up spills,
and follow all the safety recommendations on the package. Always pour
the ferric chloride dry etchant powder into the water. Do not pour the water
onto the ferric chloride. When you pour the ferric chloride into the water,
do so slowly. Use distilled water for this solution.
204
Planar Fuel Cell Stack
1
About 1/ 2# of ferric chloride at most is needed to etch both boards; /4#
is supposed to be enough to etch 200 square inches. Each board is about
80 square inches (8"x 10"), but more etchant may be required. Follow the
directions that come with etchant for mixing. One pound of etchant
requires 60 ounces of water, so for one or two boards, you need 15 ounces
1
of distilled water for each /4#
(4 ounces) of etchant powder.
Mix the crystals in with the
water and fill the tub enough so
that the solution will cover the
top of the circuit board during
the etching process.
Use only distilled and very
warm to slightly hot water to
mix with the etchant powder. A
warm etching solution works
much faster than a cold solu-
tion, so make just enough solu-
tion at one time to etch one
Bathing the plates in
board. The solution cools as it
etchant solution
205
Planar Fuel Cell Stack
sits during the etching process and thus the etching process slows down.
A heating pad turned on high placed under the etching tray can help to
maintain a good etching temperature.
To etch, slip the board into the tub filled with solution. Every few min-
utes, agitate the board by moving it around or tilting the tray back and
forth to move the solution across it. This motion is necessary and makes
the etching go much faster. As you do this, you can see the copper disap-
pearing in the areas that are not covered by the resist.
When all the copper has been removed where it is supposed to be
removed, take the board out of the solution and rinse with water. This
should leave a perfect image of the electrode grids in blue. Do not leave
the board in the etchant longer than necessary, or the solution will begin
to undercut the electrode pattern. This is not a process in which you can
go away for a while and then come back. You must monitor the whole
process, agitate the tray and immediately remove each board when the
process is complete.
206
Planar Fuel Cell Stack
The electrode plates with the etching process
complete. Left, oxygen; right, hydrogen
Remove the resist
When the etching is complete, remove the blue colored resist by scrub-
bing it off with a Scotch Brite® type pad under running water. As you
scrub, the copper trace will reveal itself. Remove all the resist and the
plates will be done.
These two plates are the positive and negative electrodes and gas flow
fields for the membrane electrode assembly.
207
Planar Fuel Cell Stack
Routing the flow fields
The flow fields can be routed either on a drill press or a small hobby milling
machine with a 1/8 router bit. If you use a drill press, attach a table to set up
a fence, so that you can move the circuit board against the fence as you rout
the gas slots.
It's a good idea to purchase extra circuit board to practice on to get used to
milling this particular material. It is easy to work with, but getting a feel for it
before you work on the actual piece for the fuel cell can be very helpful.
If you don t have a drill press or milling machine, you could ask a friend, or
some enterprising high school shop student to do the job; or take a class at a
local tech school and bring your circuit board with you as a project. Another
option is to take the circuit board to your local machine shop and have them
rout the slots for you.
Whatever you do, get extra circuit board for either you or someone else to
practice on before the final pieces are cut. Even the most experienced
machinist may not have any experience working with this material.
208
Planar Fuel Cell Stack
Because of the nature of the thin copper cladding, be sure that the router
1
does not gouge the material. The /8" router bit (#60719 from Micro Mark) I
have indicated seems to work best, although other kinds will work. Be sure to
supply this router bit to anyone who does the job for you unless they are con-
fident that the router bit they will use won t make the copper edges too ragged.
Using a drill press or milling machine
If you use a drill press, I suggest using an XY table. With an XY table you
can just clamp the piece to the table and turn a wheel which moves the piece
through the cut while you hold the spindle down. If you do not have an XY
table, use a fence, and guide the circuit board through the cut with your
hands, holding the piece securely. You will need to have someone else lower
and raise the spindle for the cuts as you guide and hold down the work
piece, so that it does not vibrate or pull up during cutting. When cutting the
circuit board material, move the piece slowly through the cut with a little
pressure and let the router do most of the work.
I have done this both ways and I can assure you it is a lot easier with an
XY table on a drill press, and easiest with a milling machine. Either type of
machine must have enough throat depth to move the 8"x 10" piece around
and position it to complete all the cuts. Be sure to check this out  some
machines do not have enough throat depth for this project.
209
Planar Fuel Cell Stack
Setting up
Most of the work in milling is setting up, that is, planning how to make
the cuts. Before you start routing the flow fields, know exactly how you
will proceed to the very end, and the milling itself will be an easy task.
Place a piece of Plexiglass® between the circuit board and the table to
act as a cutting buffer, and use some small pieces of Plexiglass® on top
of the piece to protect the its surface from the clamps, if you use clamps
(see below).
A milling machine
set up to mill
the oxygen
electrode plate
210
Planar Fuel Cell Stack
Controlling the depth
Rout all the way through the circuit board. The router bit must be set at a
depth that is just a little more than the depth of the circuit board to make a
complete cut. If you use a drill press, there are two adjustment nuts on a
screw that act as a drill stop or depth stop. Again, set them for the bit to just
pass through the circuit board and into the Plexiglass® a little bit. If you use
a milling machine, there is also a drill depth stop that must be set before
milling proceeds.
When everything is set up, simply pull the lever down and the router will
drill into the board. With a milling machine, the router can be locked into
position, but with a drill press you must keep pressure on the handle to
make sure it stays down, as there is a spring which brings it back up if you
do not hold it.
If you are using a drill press without an XY table, this is when you need to
have someone hold the lever down while you guide the piece with your
1
hands to route the slots. To rout, simply line up the router bit on the /8"
spaces between the copper traces of the electrode fingers, set up the fence
for a cutting guide and push the piece (on a drill press without XY table)
along the slot space.
211
Planar Fuel Cell Stack
If you are have an XY table on a drill press or a milling machine, simply
turn the wheel to move the clamped piece through each cut. At the end of
each slot, lift the router from the piece. Align the piece for the next cut and
reposition the fence, if you're using a fence; or if you have an XY table,
move the wheel to reposition the piece for the next slot cut.
With smaller milling machines, and drill presses with an XY table, you
still have to occasionally reposition the piece and reclamp to make all the
cuts. If you use something larger than a hobby milling machine or small
drill press, you may not have to reposition your work at all.
When working with the hydrogen side plate, you can move from one cut
to another within each electrode because they are connected by 90°
angles. You do not have to lift the router out, just change direction. For a
drill press without an XY table you will need two fences; or if you only have
one fence set up, stop the machine and reset the fence to move the work
piece in another direction. With a fence on a drill press, it may be more
convenient to set up the fence on one side of the press, drill all the verti-
cal slots on the piece, then change the fence and drill all the horizontal
slots on the hydrogen side. Remember to rout the slots for the screws and
the edge connectors.
Experiment on your test pieces to see what works best for you.
212
Planar Fuel Cell Stack
The plates after milling. Oxygen, left; hydrogen, right
Smooth the edges and clean the plates
After routing the gas flow fields, smooth all the copper edges of the
electrode fingers with a very fine file, and/or cut a piece of 1500 grit sand-
paper and smooth the edges. It is important to have smooth edges. Parts
of the copper trace will be laid against membranes, and rough edges on
the flow fields can scratch or puncture a membrane. After smoothing the
edges, clean the copper traces again using a fine Scotch Brite® type pad
213
and then rinse with water.
Planar Fuel Cell Stack
Materials for plating the circuit
Copper corrodes quite readily and under fuel cell conditions can deteriorate
rapidly. To ensure that the surface remains conductive, a plating coat of a less
reactive metal is recommended. There are several options for coating the cop-
per surface to avoid losing the capacity of the cell. One is to plate the copper
surface with Tinnit"!. This is a tin base that will protect the copper coating.
Another option is to plate the copper surface with nickel; or, plate the cop-
per surface with nickel and then over plate it with a thin layer of gold.
The goal is to have as little resistance in the circuit as possible so that you
can get as much current from the cell as possible. Copper is an excellent
conductor of electricity with a low resistance of about 1.7 micro ohms-cen-
timeter at 20°C. Unfortunately it oxidizes and tarnishes, and this deteriora-
tion will impede the power output of a fuel cell after a while. Gold does not
oxidize readily and is quite conductive, with a resistance of about 2.4 micro
ohms-centimeter at 20°C. Even though it does not have as high conductivity
as copper, gold will survive fuel cell conditions for a longer period of time.
Nickel is highly resistant to corrosion and is an excellent metal to use, but it
has a higher resistance (about 7.8 micro ohms-centimeter). Another option
214
is tin plating which has a resistance of about 11.5 micro ohms-centimeter.
Planar Fuel Cell Stack
So, tin plate, nickel plate, or nickel plate with
additional gold plate are acceptable options,
but the most desirable is the nickel-gold plate.
The straight nickel plate would be the second
best and the tin plate would be the third choice.
For this project we decided to put a nickel strike
on the copper, and over plate it with gold to
enhance conductivity. This took more work and
was a little more expensive, but it enhanced the
performance significantly so it was worth the effort.
Brush plating
The easiest method to plate these circuits is
Applying nickel plating
brush plating. Brush plating is simple electro-
over the copper layer
plating with a brush plating wand. The wand
has an absorptive tip which is dipped in the
plating solution, and then rubbed and stroked onto the surface to be plat-
ed. The pen or wand is connected to the positive pole of a battery or power
supply, and the work piece to be plated is connected to the negative pole
of the battery or power supply.
215
Planar Fuel Cell Stack
As the pen is gently rubbed on the
object to be plated, a thin film of
metal is deposited on the surface. If
you plate the electrode surfaces with
gold, plate them with nickel first. If
you simply plate the copper with
gold, the gold is absorbed or
migrates into the copper over a short
period of time. To prevent this, a
nickel plate base is necessary.
Plating kits
Applying gold plating
If you have never brush plated over nickel plating
before, you may wish to purchase a
inexpensive brush plating kit from Caswell plating. They have Plug N
Plate® kits which come with a plug in power supply (a simple 300 milliamp
wall transformer), a brush plating wand, plating solution, and instructions.
They have a nickel Plug N Plate® kit which includes plenty of nickel
solution to cover the plates. If you choose to use gold also, purchase a
small bottle or two of gold plating solution. The nickel plating kit wand and
power supply will work with the gold solution. An added benefit is that if
216
you order the kit, you can also get technical support for using it.
Planar Fuel Cell Stack
To apply tin plating, you can use a Plug N Play® kit, or use an electroless
immersion plating solution such as Tinnit"!. You could also have a plating
shop do the plating.
Every part of the metal circuit should be plated except for the tips of the
series edge connectors. These will be tinned to be soldered with a solder-
ing iron. If you use Tinnit® for plating, you can plate the tips also. Other met-
als require different types of solders, so it is best to leave the tips unplated
where you will solder the tabs to ensure good connections.
Tinning the series edge connectors
The next step is to tin the tips of the series edge connectors on the oxy-
gen and hydrogen electrode/flow field plates. There are twelve contacts to
be tinned on each plate, eleven for the series connections and one for the
take off on one end of each plate.
To tin, take a 60 watt soldering iron and touch it to the solder.
The solder will stick to the surface of the iron tip. Then, rub the sol-
der off the tip onto the connector.
Tin the edge connectors
Preparing the tab connects
on each of the
electrode plates.
To connect the cells in a series,
217
use tab wire. You can either purchase
Planar Fuel Cell Stack
the tab wire or cut it from copper foil. Use any
thickness you like, but it should be flexible
enough to work well with this circuit. Generally,
a .005 thickness is sufficient. Cut eleven tabs
1
/4" long, or,a tad longer as there is a small sur-
face bend to accommodate between the two
plates. Don't cut the tabs too long, but don't cut
them too short because you want as much con-
tact surface as possible. Tin the tabs on both
Solder 1/4 " tabs to the
sides. Solder the eleven series connect tabs to
edge connectors on
one of the plates.
one of the plates.
Left, after
soldering
the tabs to
the plate,
test for
continuity.
218
Planar Fuel Cell Stack
Making the electrode gasket
From either .010 or .020 silicone rubber, cut
the hydrogen gasket with an Exacto® razor
knife using the template (see page 248) as a
guide. This thin silicone material comes sand-
wiched between two pieces of Mylar®. The
Mylar® will be used to mount the MEAs (mem-
brane electrode assemblies). When you
remove the silicone from the Mylar®, the sili-
Preparing the silicone
cone will shrink, so do not cut it while it is
hydrogen gasket
for the MEA
219
Planar Fuel Cell Stack
sandwiched between the Mylar®. Remove it
from the Mylar® sheets and let it sit for a few
minutes to shrink to its working size.
Making the surrounds
Cut the two Mylar® sheets using Template 3
as a guide. It is very easy to over cut when
using the Exacto knife, so go slowly, and use
fresh blades to be sure to produce clean cuts.
MEA (membrane
electrode assembly)
Inserting the membranes
Exercise care when handling the MEAs. It s a
good idea to wear cotton gloves so that you do
not get any oil, sweat or other contaminants on
the membrane and catalyst. If you purchase
MEAs from Element 1, the anode side is marked
as you can see in the photo. All anodes must
face the hydrogen electrode plate.
Each MEA needs to be glued and sealed onto
the surface of the Mylar® surrounds. A good
220
seal must be formed when attaching each MEA
Planar Fuel Cell Stack
so that gases from one side
cannot travel to the other
side. A bad seal will result in
a non-functioning fuel cell.
Before gluing the MEAs,
center each of the MEAs
over the holes they will be
mounted in and observe
whether the ionomer of any
membrane touches the ones
next to it. Be sure that the
Trim the ionomer so that the individual
MEAs do not touch any of the others
ionomers do not touch. If they
do, trim them down so that
they do not come into contact with each other. Trimming will negate the pos-
sibility of ionic cross conduction in the membrane, which can reduce the ener-
gy output of the fuel cell stack.
Although the individual membranes must not touch each other, do not trim
off too much of the ionomer because there should be as much contact sur-
face for gluing as possible. The larger the contact gluing surface, the more
likely the membrane will be properly sealed.
221
Planar Fuel Cell Stack
Once the spacing has been fine tuned, the membranes can be glued. Lay
one of the Mylar® sheets in front of you and apply silicone adhesive to the
glue area on the Mylar® for the first membrane to be glued. Pick up the
membrane and very carefully center and place it, being sure not to get any
adhesive on the gas diffusion layer (the dark colored active surface area).
Press and smooth the ionomer to the Mylar® surface so that it is well
secured and soundly attached. Make sure to use enough silicone to cover
the glue area very well, but not so much that it will ooze out beyond the glue
zone. This needs to be a clean job. If any silicone bumps are left and they
dry, they will interfere with getting a good gas seal. Make everything as
clean and smooth as possible.
When you press the ionomer to the Mylar®, some of the silicone will ooze
out on the edges. Smooth it out immediately, because once it begins to hard-
en, it will be extremely difficult to remove it without damaging the membrane.
After the surface facing you is clean and smooth, turn the Mylar® sheet
around and smooth out any oozing surface on the other side. When you turn
it around, never place it on the same piece of paper as you might acciden-
tally get some of the silicone on the active surface of the MEA. Use a fresh
sheet of paper each time you turn the piece. You will go through a lot of
paper but the MEAs will come out unscathed. If any part of the active area
222
Planar Fuel Cell Stack
of the MEA is covered, it will reduce
the output of that cell, and in a series
connected stack like this it will also
reduce the output of the other cells.
After completing the insertion of the
first MEA, complete the other eleven.
To complete the MEA sandwich,
cover the surface of the other sheet of
Mylar® with a smooth and complete
coat of silicone to adhere it to the first
piece of Mylar®. Very carefully lay the
second sheet over the Mylar® mem-
brane assembly in front of you. Press
and smooth the two pieces together to
ensure a gas tight seal.
Put a smooth layer of silicone
around each MEA where the Mylar®
The completed MEA sandwich:
edge touches the membrane to further
12 MEAs mounted between
seal the edges against gas leakage.
two sheets of Mylar®
223
Planar Fuel Cell Stack
Do this on all four sides of each of the 12 membranes. Set this assembly
aside to dry, placing it so that it will not stick to any surface.
Gas supply lines
Use Template 4 (page 248) to cut the rubber gas
3
supply lines from the /32" thick silicone rubber
sheet. Cut the rubber to the correct size and then
tape the template to the rubber. Use an Exacto®
knife or other very sharp cutter to make the gas
lines. I used a small Exacto® razor knife and found
that I had to change blades very frequently  I ran
through a whole pack cutting out the gas lines. Cut
slowly, and when the template paper starts to
Above, cut the gas
bunch up as you are cutting, stop immediately and
supply lines out of
change the blade. If the template paper rips you
the rubber with
will lose the outline which will make it difficult to cut
an Exacto® knife.
accurately. The process is easy, but as with other
operations in this project, it can be tedious.
Attach the gas supply gasket
Lay the hydrogen electrode plate on your work surface with the metal
electrode side down. Lay the gas supply gasket next to it with the side of
224
Planar Fuel Cell Stack
the gasket that will be glued to the hydrogen
plate facing up.
Coat the gas supply gasket with a thin layer of
silicone caulking. Pick up the
gasket and position it quickly
onto the hydrogen plate
before the silicone gets too
tacky and sets. All of the gas
supply lines must be aligned
The rubber gasket laid
so that they feed into and out
and glued on the
of each cell properly. Check
back of the hydrogen
electrode plate.
and adjust these quickly
before the silicone caulking
sets. Press and rub the rubber gasket to make sure the
gasket is sealed against the electrode plate surface. Set it
all aside and let it dry for twenty four hours.
Note the illustration at left, which shows the hydrogen
electrode with the gas feed gasket glued to its back, as
seen from the electrode side. The orange gasket is visi-
225
Hydrogen electrode plate with orange rubber
gas feed gasket glued to its back.
Planar Fuel Cell Stack
ble through the routed flow fields in the circuit board, and also somewhat
visible through the unplated parts of the board. Also note the gas inlet and
outlet holes at the beginning and end of each flow field.
Preparing the end plate
Draw an 8"x 10" outline centered in the PVC
plate. Align a copy of Template 5 (page 248) for
the Pressure Plate inside this 8"x 10" outline.
Take a sharp object and press the point into the
cross hairs of the 1/4" and 1/2" outlet, and the six
holes for the pressure screws to make indenta-
tions to mark the PVC plate.
Remove the template, and then drill pilot
holes for the gas inlet and outlet with a small
The PVC end plate with
size drill; then drill the final holes with the cor-
holes drilled for the gas
rect size drills. The pilot holes are not neces- inlet and outlet and the
pressure screws
sary, but they can help your drilling to be more
accurate. Also, do not drill the final hole too big.
There should be as tight a fit as manageable for
the barb hosed connector.
226
Drill the six 1/8" holes for the pressure screws.
Planar Fuel Cell Stack
Aligning the plates
You can use the PVC end plate as a
base to align the electrode layers. To do
this, three of the the pressure screws
are inserted through their holes in the
end plate, but in the opposite direction
of the way they will be installed when the
fuel cell is completed.
Put the hydrogen electrode plate (with
Three pressure screws are inserted
the gasket glued to it) on the PVC plate
through the PVC end plate to
with the metal electrode surface up, and
align the electrode layers.
th gasket side down; and align it with the
guide screws. Wash the .010 or .020 sil-
icone gasket material with water and dry
it, then lay it on the hydrogen plate. Align
it with the electrode cells (see illustra-
tion, top right).
Place and align the Mylar® MEA on
top of the silicone electrode gasket
(see right).
227
Planar Fuel Cell Stack
Then, as shown at right, take the
oxygen side plate and place it on top
of the Mylar® MEA, with the plated
electrode surface face down. and
align. Add the other three screws
and tighten them down slightly with
butterfly nuts, as shown in the photo.
They should be tight enough to hold
the layers in place, but should not
Ready to solder with all the
exert downward pressure on the lay-
electrode layers in place.
ers, as this could interfere with sol-
dering the connectors.
Soldering the
connectors together
Soldering the series connections
and power takeoff tabs
Since the series interconnect tabs are
already soldered to the oxygen plate, all you
have to do is solder the tabs to the hydrogen
plate to complete the series connection. To
do this, put the soldering iron tip in the
space where the tab is and push down until
228
Planar Fuel Cell Stack
the solder melts and a bond is formed. A sixty watt iron with a chisel tip
usually works well and is the correct size to fit in the space allotted. This sol-
dering operation is tricky because you do not want to desolder what you
already soldered.
If you don t have much soldering experience and are not confident
about this operation, you can make tabs that extend away from the plates
and connect them together away from the fuel cell (see illustration at
1
right). Instead of using /4" tabs as shown on page 218, make the tabs a
little longer, and cut 22 of them. Then, solder a tab to each
series connection on both plates, instead of just to one
plate. After the component layers are aligned as described
before, solder the tabs together away from the edges of the
plates. This may be an easier method; however it will not be
as neat and compact looking.
Whatever method you use, be sure that each connection
is good. If one of them is not, the fuel cell won t work.
Alternative
When you have finished soldering the series connections,
way to
make the con-
the positive (oxygen) side of each cell will be connected to
nections
the negative (hydrogen) side of the next cell in series, so
the voltage from each cell is added to get the total output of
229
this planar fuel cell stack.
Planar Fuel Cell Stack
Next, cut two pieces of 1/4" wide by .005 thick tab ribbon. These tabs are
the power take off tabs. They can be cut long, and trimmed to the desired
length when the fuel cell is finally assembled. One tab is connected to the
hydrogen side take off connector and the other to the oxygen side edge take
off connector. You will be able to solder one of the takeoff tabs with the com-
pression screws on, but the other takeoff is on the other side and you will
need to solder that when you take the sandwich off the PVC plate and turn
it around. Tin the tabs before soldering. Cover a portion of the power take
off tabs with a liquid electrical tape to insulate them from touching the com-
pression screws and aluminum channel bars.
Carefully loosen the compression screws and gently remove the sand-
wiched layers from the PVC plate, solder the other takeoff tab, and set
them aside. The sandwich may tend to fan out on the edge opposite the
edge with the soldered connections. This is not a problem and is to be
expected. When the sandwich is finally laid in between the PVC plate and
the channel holds, and the edges are sealed, the layers will be evenly
compressed.
230
Planar Fuel Cell Stack
Preparing the pressure braces
Pressure braces ensure even pressure across the electrodes so that
they connect well with the MEAs.
Aluminum is readily available and inexpensive, but aluminum in bar or
rod shape has too much bend for this application. We chose aluminum
channel for the pressure braces. Because of its structural shape, the chan-
nel stock has stiffness that flat pieces of thicker metal bar do not have.
5 5
Cut three 12" pieces of /8"x /8" aluminum u-channel for pressure
1
braces. Drill /8" holes on either side of the channel to match the holes in
the PVC plate so that they mate correctly.
Debur all the edges and then wash the channel to remove any aluminum
powder from the cutting, drilling, and deburring process. Any metal powder
that later finds its way onto the membrane surface via the air ports could
231
be detrimental to the operation of the cell.
Planar Fuel Cell Stack
Installing the gas port connectors
Prepare the barb connectors for installation in the
gas port holes in the PVC plate. Cut the connectors so
that they will be flush with the inside of the PVC plate.
Apply epoxy glue to the inside rim of the drilled holes
in the plate, then apply epoxy to the outside rim of the
Barb connectors
barb connectors that will contact the PVC plate. Insert
for gas ports
the barb connectors into the holes and then apply
more epoxy to the top and bottom edges where the
barbs contact with the PVC plate to assure a good seal. Clean off any
excess epoxy, but also leave some to provide ample seals. Avoid getting
any epoxy inside the barb connector. This would block the gas flow
through the connector. Lay the PVC plate aside and let it dry.
Final assembly
Place the electrode assembly onto the PVC plate and align. Take a
piece of channel and position it, insert the screws in the holes with the
screw heads inside the channel. On the opposite side of the plate, place
a washer around the screw, put a wing nut or other nut onto the screw and
twist it down to hold the channel and the plate together. Finish by apply-
232
ing the other two pieces of channel to the plate.
Planar Fuel Cell Stack
Tighten the wing nuts by hand, and
do not let the plate bend from over
tightening. The plate should be
straight across without the slightest
induced curves.
Keep the take off tabs clear of the
screws as you insert the screws. In
this design, one of the take off tabs
comes very close to one of the
screws, so it is important to move it
around the screw if necessary and be
sure that the tab is coated with an
insulated material such as silicone,
liquid electrical tape or rubber dip
Completed fuel cell, with the back side
compound as mentioned earlier to
of the oxygen plate visible
prevent contact between the ribbon
and the screw or channel. For more rigidity you may want to insert stiff
metal strips under the channel at several points to insure more even com-
pression. If you do this, make sure you do not cover the air slots on the oxy-
gen side with the metal strips.
233
Planar Fuel Cell Stack
Apply a coat of silicone to the
edges of the electrode assembly.
Completely cover the edges and
where they meet the end plate to
ensure a gas tight seal. Then,
smooth the edges with your finger or
a razor blade to make a nice clean,
even looking finish. Set the fuel cell
aside to let the silicone dry.
Prepare the stack for testing
Inspect the fuel cell stack to see
that everything is in order and the
Completed fuel cell showing the
pressure screws are tightened. If you
hydrogen inlet and outlet
have doubts about the silicone cover-
age on the edges, apply more to
ensure a tight sea, then let it dry for 24 hours before testing.
You can attach a small piece of hose to the gas exit port barb connec-
tor and connect a valve to control the amount of gas that exits the cell.
This exit port gas regulation can help to fine tune and tweak the cell out-
234
put. You should be able to "dead end," that is, go from a completely closed
Planar Fuel Cell Stack
port and open the valve in increments for more exit volume. Please note
that you should not dead end the cell if there is no storage container with
pressure relief as outlined in this book, or some such other exit for the
gas. The GPS system has a check valve in the design which prevents the
gas from flowing backwards in the system. This can cause excessive pres-
sure build up in the fuel cell stack which can cause gas leakage, or rup-
ture the membranes which will destroy the stack.
Purging the stack
To purge the stack of air before introducing hydrogen you can use bot-
tled gas, or purchase a good size helium balloon. When the balloon is
filled, have it left tied loosely so that you can open it easily. Put the noz-
zle of the balloon on the exit port of the fuel cell and let the gas expel into
the fuel cell and out the entry port. At the same time, have the regulator
turned on to provide hydrogen and insert the hydrogen hose onto the
intake of the fuel cell. Remove the balloon and the stack
is purged.
We have not used inert gas purging, since our system is experimental
and set up for short runs. For a permanent system, if there is a need to
purge the stacks, inert gas purging and a built in purging system should
235
be considered.
Planar Fuel Cell Stack
Testing the fuel cell stack
To test the fuel cell, your hydrogen production system needs to be set
up and running, or you need another source of hydrogen available.
First, connect the gas supply line from the regulator to the fuel cell entry
port barb connector. As a general practice, always have some sort of
flashback arrestor between the regulator and the fuel cell stack. This can
be a small flashback bubbler that you can construct and install, or you can
use a commercial flash arrestor.
Use a multimeter or voltmeter to do an open circuit voltage test.
Connect the leads from the meter to the positive and negative take off ter-
minal tabs on the fuel cell.
If you are using the GPS system design in this book, open the regulator
that is connected to the electrolyzer/storage system to allow hydrogen gas
to flow to the fuel cell. If there is a gas exit valve on the fuel cell, be sure
to open it for full gas exit. Then, you can diminish the gas flow in incre-
ments if you wish during testing.
Allow the gas to flow to the fuel cell for a minute or two, and then take
the reading from your voltmeter. Readings should be from 15V through
9.6V. After a while the voltage will drop and plateau at various points. You
236
Planar Fuel Cell Stack
can expect the cells to settle out and maintain a constant voltage of
around 7.5 to 10.8 volts. This will vary, however and you will have to test
your particular fuel cell stack to determine its characteristics.
This small power stack was designed to produce 6V at between 1 to 2
amps under load. Much of the current output depends on the quality of
the membranes used. If you purchase untested membranes from Element
1, expect anywhere from .5 to 1 amp or more output.
To test the short circuit current simply change the multimeter setting to
the current reading or use an ammeter and connect the probes to the
same terminals that used to get the voltage reading.
Trouble shooting
If you do not get any readings, you have just entered fuel cell builder's
hell. This is a place where your shoddy previous work (or somebody
else s) comes back to haunt you, and you are forced to contemplate and
work to fix your mistakes.
So, where could the problem be and what is the best way to trou-
bleshoot your system? The first rule of troubleshooting is to prevent trou-
ble in the first place as much possible. Do not rush while constructing the
stack. Test, check and recheck as you go along, and do things right the
237
Planar Fuel Cell Stack
first time so that you don t have to troubleshoot. Then, if there are still
problems, you can be confident enough to rule out many possible causes
of trouble, and zero in on the real cause and fix it.
Make a troubleshooting check list to have in hand before testing the fuel
cell stack, and follow the check list regimen always, because if you don t,
it is very easy to get confused.
In a system of this nature there can be many different problems. First,
check for the obvious. Are the multimeter probes connected? Is the multi-
meter working properly?
Then, look at the system as a whole. A simple visual inspection of a sys-
tem can sometimes show the problem immediately. Is everything connect-
ed properly? Are there any gas leaks causing pressure failure or loss of
gas? Are valves connected properly, and turned on or off, at whatever set-
ting they should be for the system to work?
If you are sure the system is getting gas, then consider closing the valve
in increments on the outlet of the fuel cell to see if that produces any
response on the multimeter. Do this until it is dead ended. After a short
period of time, if there is no meter response, open up the valve again and
move on to another test.
238
Planar Fuel Cell Stack
Next check the fuel cell stack itself, and whether or not there is gas
leaking between the plates, etc. If there is a leak here, it must be
resealed.
Since this stack is connected electrically in series, if one of the
PEMbranes is not working, it will shut down the whole stack. This is why
each membrane should be tested prior to installation in the stack.
However, there can also be damage to a PEMbrane during construction of
the fuel cell. Another possibility is that the connections are not soldered
well and you will have to inspect the tabs to see if that is the case. More
than likely there will be no problems.
To test individual PEMbranes prior to assembling them in the final MEA,
you can make a small test fixture. Construct a single slice fuel cell, such
as the easily constructed L78 detailed in Build Your Own Fuel Cells. To
use the L78 as a test cell, don't glue the Mylar® sheets together, just
insert the PEMbrane between the two Mylar® sheets along with the gas-
kets and other layers of the fuel cell, and hold the layers together for the
test with insulated binder clips, or use screws as outlined in the text. You
can of course, improvise on this as you wish.
239
Planar Fuel Cell Stack
Designing fuel cell stacks
The L79 is a planar variation of some of the construction ideas involved
in the L78 stack, as presented in Build Your Own Fuel Cells, which also
has printed circuit board for the electrode plates. While working with this
material, we realized that flow fields could be routed in the circuit board,
so that the material could function for a dual purpose: as both a collector
plate, and a gas flow field.
As you construct the L79, no doubt many new ideas and configurations
will come to mind. After completing this fuel cell stack, you should feel
confident enough to make modifications to improve the performance; and
you'll probably discover better or more inexpensive materials and methods
for fuel cell construction. Once you understand the basics of fuel cell con-
struction, it is easy to come up with many variations.
Fuel cell power supplies
You may wish to build a fuel cell stack power supply with greater volt-
age and current output.
A good start is to construct a 12 volt 10 ampere power supply, which
can be constructed with either planar or block stacks, as discussed in
Build Your Own Fuel Cells. Whatever design you choose, this project will
240
Planar Fuel Cell Stack
require 120 MEAs that will deliver around .5 volt and 2 amperes each
under load.
Shop around to get the best price for MEAs. There are URLs for sever-
al suppliers that can be accessed via the Resources section of this book.
In the process of talking with suppliers you will learn a lot about MEAs,
which will help you decide which product is best for you. You can also
make your own MEAs, and if this interests you, Build Your Own Fuel
Cells has a detailed section about this subject.
To construct the power supply based on the L79 planar stack will require
ten stacks of twelve MEAs each, and will have an output of 6 volts at 2
amperes per stack. Five series connected pairs of stacks produce 12 volts
at 2 amperes per pair. The five pairs are then connected in parallel for a
total output of 12 volts at 10 amperes for the power supply (see below).
Wiring for 5
pairs of series
connected
stacks, wired
together in
parallel.
241
Planar Fuel Cell Stack
These planar stacks can be racked in a variety of ways. The above illus-
tration shows a simple tubing manifold system for feed lines and gas exit
lines. On the next page is another view of the rack of stacks.
In designing a power supply, consider the number of photovoltaic pan-
els needed, as well as how many electrolyzers. Then, size the gas pro-
cessing system and storage to provide the appropriate quantity and qual-
242
ity of gas for your fuel cells.
Planar Fuel Cell Stack
Rack of planar fuel cell stacks
243
Planar Fuel Cell Stack
Output configurations
You can take power straight from the fuel cells; or use supercapicitors to
condition the output; or feed small rechargeable batteries if you have spe-
cial applications that require them.
For instance, if there are high amp draws from motor start ups, etc., put
a supercapacitor in parallel connection with a 12 volt rechargeable battery,
and use this to supply those intermittent load needs adequately. To use a
rechargeable battery alone, as mentioned, simply connect the output from
the fuel cells to the battery and draw power from the battery. To use a
supercapacitor and rechargeable battery, connect the battery and superca-
pacitor in parallel, connect the fuel cell output to these, and draw your
power from the supercapacitor and battery connected leads. With these
system additions you will need a diode so that reverse flow does not occur
to the fuel cell stack, and fuse the circuit on both sides in case of shorts. A
switch, either remote or direct, should be used to connect the power supply
with any lines or equipment being powered. If you have AC power require-
ments you will need an inverter to convert DC to AC electricity.
244
Planar Fuel Cell Stack
Running the stacks
In operation, let the fuel cells run for a few minutes before connecting
to a load. It would be a good idea to install a voltmeter and ammeter from
the fuel cell to output line, and also at other junctions if so desired, so that
you can monitor the system at various points. These systems can be total-
ly automated, manually controlled, or both.
Cold weather can wreak havoc with PEM fuel cell stacks for a variety of
reasons. It is important to set up the power supply so that it is as well
insulated as possible. At the same time, proper air flow must be main-
tained to feed the cells, and adequate ventilation provided to keep any
hydrogen gas from accumulating.
Hydrogen gas monitors that are correctly calibrated on a proper sched-
ule should be used in any power supply system to provide alarms for haz-
ardous conditions. Automatic cut offs can be used in conjunction with an
alarm system to shut down the gas flow to the power supply.
System maintenance and reliability are very important, so be sure to
consider this in every step of your design. Engineering a practical and
safe fuel cell power supply takes a bit of thought. It is best to start out
small and build up slowly, so that your mistakes do not turn out to be cost-
245
ly or dangerous mistakes.
Planar Fuel Cell Stack
Of course, you can purchase a stack and then build the rest of the sys-
tem around that, or purchase other system components as well. Fuel Cells
2000 has an excellent web site with a list of available fuel cell units with
photos, specifications, and prices (access URL via Resources page).
Stacks can also be purchased on the surplus sales market.
At the time of this writing, new 1KW stacks can be purchased for around
$6000.00 to $8000.00. Used units of the same output may cost around
$4000.00 to $6000.00 Most current stacks have an operating life of 1500
to 2000 hours. More expensive units can go above the 2000 hour range.
It may seem at first that building a power supply is more expensive
than buying one. However, the average life of these purchased units is
only 12 to 24 weeks of continuous duty, so replacement costs and serv-
icing fees can make building your own seem more reasonable. If you do
buy a unit, find out how easy it will be to replace the parts and perform
other maintenance.
Have fun, and always be safety conscious, so that you can have many
more days of experimenting and discovery.
246
Resources
Safety - MSDS and hydrogen safety guidelines
Hydrogen Resources
Supplier URLS (Click here for a more extensive list)
McMaster-Carr  Many, many useful items
HMC Electronics  Rosin, flux pens, solder
E. Jordan Brookes Co., Inc.  Tab and bus ribbon
Plastecs  Solar cells, tab ribbon
All Electronics  Electronic and electrical supplies
Surplus Sales Of Nebraska  Electronic and electrical supplies
Element 1 Power Systems  MEAs
Ocean State Electronics  Printed circuit board, ferric chloride
etchant, trays, tongs,
Micro-Mark  Hobby milling machines, routers, end mills, and lots
of other items useful for fuel cell building.
Caswell Plating  Electroplating and electroless plating supplies.
Techniks Inc  Press-N-Peel PCB Transfer Film
247
Templates for the L79 Fuel Cell Stack
If possible, set your printer options to print to the edge of the paper, or print the
templates to a paper size such as 11 x 17 , if your printer has that capability.
Because the electrodes and MEA assembly are 8 wide, some printers may not be
able to print either the right or left edge for Templates 1, 2, 3, or 5. This will not
matter for printing to the transfer film for the electrode templates  they have been
aligned so that all of the image needed for the transfer can be printed by any laser
printer on the 8.5 x 11 transfer film sheets. For the other templates, if an edge
line is missing, you may wish to take a ruler and mark it on the template at exact-
ly 8 wide. Use the on-screen template page as a reference.
Also, for printing to the transfer film, adjust the laser printer settings for maxi-
mum coverage, or the darkest/highest quality option for printing.
Click the red text to open the template page. To close the template page
and return to this page, click the cross in the upper left hand corner.
Template 1  Negative (Hydrogen) Electrode Plate, from PC board
Template 2  Positive (Oxygen) Electrode Plate, from PC board
Template 3  MEA Surrounds and Gasket, 2 mylar and 1 silicone rubber
Template 4  Rubber Hydrogen Gasket, silicone rubber
Template 5  PVC Pressure Plate
248
Other e-book titles available from
Wheelock Mountain Publications:
Build Your Own Solar Panel
by Phillip Hurley
Build Your Own Fuel Cells
by Phillip Hurley
Solar Hydrogen Chronicles
edited by Walt Pyle
Tesla: the Lost Inventions
by George Trinkaus
Tesla Coil
by George Trinkaus
Radio Tesla
by George Trinkaus
Chinese Firecracker Art
by Hal Kantrud
Wheelock Mountain Publications
is an imprint of
Good Idea Creative Services
249
324 Minister Hill Road
Wheelock VT 05851
TEMPLATE 1 - Negative (Hydrogen) Electrode Plate
TEMPLATE 2 - Positive (Oxygen) Electrode Plate
TEMPLATE 3 - MEA Surrounds and Gasket
Cut 2 from mylar, and 1 from .010 or .020 silicone rubber
TEMPLATE 4 - for 3/32 Rubber Hydrogen Gasket
gas outlet and inlet positions
(for reference only)
TEMPLATE 5 - for PVC Pressure Plate
Use this template to mark the screw holes
and gas inlet and outlet on the side
of the PVC Plate that will face the
rubber gas gasket. center the
outline on the 12 by 12 plate.
gas outlet
gas inlet